US20080039086A1

Movatterモバイル変換

Info

Publication number: US20080039086A1
Application number: US11/778,040
Authority: US
Inventors: Michael Gallagher; Milan Markovic; Patrick Tao; Amit Khetawat
Original assignee: Individual
Current assignee: Kineto Wireless Inc
Priority date: 2006-07-14
Filing date: 2007-07-14
Publication date: 2008-02-14
Also published as: US20080305793A1; US20080181204A1; US20080123596A1; US20080137612A1; US8005076B2; US20080132224A1; US20080130564A1

Abstract

Some embodiments provide a method of registering a user equipment (UE) in a communication system that includes a licensed wireless communication system and a generic access network (GAN) that has a generic access network controller (GANC). The method sends a register request message from the UE to the GANC that indicates a GAN mode capability of A/Gb only for the UE. When the GANC has a GAN mode capability of A/Gb, the GANC registers the UE with the GAN. When the GANC has a GAN mode capability of Iu only, the GANC rejects the register request message. When the GANC has a GAN mode capability of both A/Gb and Iu, the GANC registers the UE based on a set of GANC mode selection rules that the GANC applies for registering UEs with the GAN.

Description

CLAIM OF BENEFIT TO RELATED APPLICATIONS

This application claims the benefit of U.S. Provisional Application 60/807,470, entitled “E-UMA Technology,” filed Jul. 14, 2006; U.S. Provisional Application 60/823,092, entitled “Generic Access to the Iu Interface,” filed Aug. 21, 2006; U.S. Provisional Application 60/862,564, entitled “E-UMA—Generic Access to the Iu Interface,” filed Oct. 23, 2006, and U.S. Provisional Application 60/949,826, entitled “Generic Access to the Iu Interface,” filed Jul. 13, 2007. The contents of each of these four provisional applications are hereby incorporated by reference.

FIELD OF THE INVENTION

The field of invention relates generally to telecommunications. More particularly, this invention relates to a mechanism for extending Unlicensed Mobile Access (UMA) or Generic Access Network (GAN) to inter-work with a GSM core network using the Universal Mobile Telecommunication System (UMTS) Iu interface.

BACKGROUND OF THE INVENTION

Licensed wireless systems provide mobile wireless communications to individuals using wireless transceivers. Licensed wireless systems refer to public cellular telephone systems and/or Personal Communication Services (PCS) telephone systems. Wireless transceivers include cellular telephones, PCS telephones, wireless-enabled personal digital assistants, wireless modems, and the like.

Licensed wireless systems utilize wireless signal frequencies that are licensed from governments. Large fees are paid for access to these frequencies. Expensive base station (BS) equipment is used to support communications on licensed frequencies. Base stations are typically installed approximately a mile apart from one another (e.g., cellular towers in a cellular network). The wireless transport mechanisms and frequencies employed by typical licensed wireless systems limit both data transfer rates and range. As a result, the quality of service (voice quality and speed of data transfer) in licensed wireless systems is considerably inferior to the quality of service afforded by landline (wired) connections. Thus, the user of a licensed wireless system pays relatively high fees for relatively low quality service.

Landline (wired) connections are extensively deployed and generally perform at a lower cost with higher quality voice and higher speed data services. The problem with landline connections is that they constrain the mobility of a user. Traditionally, a physical connection to the landline was required.

In the past few years, the use of unlicensed wireless communication systems to facilitate mobile access to landline-based networks has seen rapid growth. For example, such unlicensed wireless systems may support wireless communication based on the IEEE 802.11a, b or g standards (WiFi), or the Bluetooth® standard. The mobility range associated with such systems is typically on the order of 100 meters or less. A typical unlicensed wireless communication system includes a base station comprising a wireless access point (AP) with a physical connection (e.g., coaxial, twisted pair, or optical cable) to a landline-based network. The AP has a RF transceiver to facilitate communication with a wireless handset that is operative within a modest distance of the AP, wherein the data transport rates supported by the WiFi and Bluetooth® standards are much higher than those supported by the aforementioned licensed wireless systems. Thus, this option provides higher quality services at a lower cost, but the services only extend a modest distance from the base station.

Currently, technology is being developed to integrate the use of licensed and unlicensed wireless systems in a seamless fashion, thus enabling a user to access, via a single handset, an unlicensed wireless system when within the range of such a system, while accessing a licensed wireless system when out of range of the unlicensed wireless system.

SUMMARY OF THE INVENTION

Some embodiments provide a method of activating a packet transport channel (PTC) in a communication system that includes a first licensed wireless communication system and a second generic access network (GAN) that has a generic access network controller (GANC). The GANC is communicatively coupled to the first communication system through a universal mobile telecommunication system (UMTS) terrestrial radio access network (UTRAN) Iu interface. The method sends a GA-PSR activate PTC request message from the GANC to a user equipment (UE). The message comprises a terminal endpoint identifier (TEID) that the GANC assigns to the UE.

Some embodiments provide a communication system that includes a first licensed wireless communication system, a second generic access network (GAN) that includes a generic access network controller (GANC). The GANC is communicatively coupled to the first communication system through a universal mobile telecommunication system (UMTS) terrestrial radio access network (UTRAN) Iu interface. The communication system also includes a user equipment (UE). The GANC includes a UDP protocol layer and a GTP-U protocol layer over the UDP protocol layer of the GANC. The UE includes a UDP protocol layer and a GTP-U protocol layer over said UDP protocol layer of the UE. The UDP protocol layer of the GANC is communicatively coupled to the UDP protocol layer of the UE. The GTP-U protocol layer of the GANC is communicatively coupled to the GTP-U protocol layer of the UE.

BRIEF DESCRIPTION OF THE DRAWINGS

The novel features of the invention are set forth in the appended claims. However, for purpose of explanation, several embodiments of the invention are set forth in the following figures.

FIG. 1 illustrates an integrated communication system (ICS) of some embodiments.

FIG. 2 illustrates several applications of an ICS in some embodiments.

FIG. 3 illustrates the overall A/Gb-mode GAN functional architecture of some embodiments.

FIG. 4 illustrates the overall Iu-mode GAN functional architecture of some embodiments.

FIG. 5 illustrates the basic elements of a Femtocell system architecture with Asynchronous Transfer Mode based Iu interfaces towards the core network in some embodiments.

FIG. 6 illustrates the basic elements of a Femtocell system architecture with an IP based Iu interface towards the core network in some embodiments.

FIG. 7 illustrates the CS domain control plane architecture of some embodiments.

FIG. 8 illustrates the CS domain control plane architecture of some embodiments.

FIG. 9 illustrates the CS domain control plane architecture of some embodiments.

FIG. 10 illustrates the UE CS domain control plane architecture of some embodiments.

FIG. 11 illustrates CS domain user plane protocol architecture of some embodiments.

FIG. 12 illustrates CS domain user plane protocol architecture of some embodiments.

FIG. 13 illustrates the UE CS domain user plane architecture of some embodiments.

FIG. 14 illustrates PS domain control plane architecture of some embodiments.

FIG. 15 illustrates PS domain control plane architecture of some embodiments.

FIG. 16 illustrates the UE PS domain control architecture of some embodiments.

FIG. 17 illustrates PS domain user plane protocol architecture of some embodiments.

FIG. 18 illustrates PS domain user plane protocol architecture of some embodiments.

FIG. 19 illustrates PS domain user plane protocol architecture of some embodiments.

FIG. 20 illustrates the UE PS domain user plane architecture of some embodiments.

FIG. 21 illustrates state diagram for generic access in the UE of some embodiments.

FIG. 22 illustrates GAN security mechanisms of some embodiments.

FIG. 23 illustrates discovery procedure of some embodiments.

FIG. 24 illustrates registration procedure of some embodiments.

FIG. 25 illustrates De-Registration initiated by the UE in some embodiments.

FIG. 26 illustrates De-Registration initiated by the GANC in some embodiments.

FIG. 27 illustrates registration Update Uplink of some embodiments.

FIG. 28 illustrates Registration Update Downlink of some embodiments.

FIG. 29 illustrates Keep Alive procedure of some embodiments.

FIG. 30 illustrates Cell Broadcast Information in some embodiments.

FIG. 31 illustrates GA-CSR Connection Establishment of some embodiments.

FIG. 32 illustrates GA-CSR Connection Release in some embodiments.

FIG. 33 illustrates Security Mode Control in some embodiments.

FIG. 34 illustrates core network to UE NAS signaling in some embodiments.

FIG. 35 illustrates UE to core network NAS signaling in some embodiments.

FIG. 36 illustrates Mobile Originated CS Call in some embodiments.

FIG. 37 illustrates Mobile Originated CS Call in some embodiments.

FIG. 38 illustrates Mobile Terminated CS Call in some embodiments.

FIG. 39 illustrates UE initiated CS Call clearing in some embodiments.

FIG. 40 illustrates CS Handover from GERAN to GAN in some embodiments.

FIG. 41 illustrates an alternative procedure performed during GERAN to GAN in some embodiments.

FIG. 42 illustrates CS Handover from UTRAN to GAN in some embodiments.

FIG. 43 illustrates an alternative procedure performed during UTRAN to GAN in these embodiments.

FIG. 44 illustrates CS Handover from GAN to GERAN in some embodiments.

FIG. 45 illustrates CS Handover from GAN to UTRAN in some embodiments.

FIG. 46 illustrates GA-PSR Connection Establishment of some embodiments.

FIG. 47 illustrates GA-PSR Connection Release in some embodiments.

FIG. 48 illustrates the message flow for PS security mode control in some embodiments.

FIG. 49 illustrates core network to user equipment PS NAS signaling in some embodiments.

FIG. 50 illustrates user equipment to core network NAS signaling in some embodiments.

FIG. 51 illustrates PTC initial activation in some embodiments.

FIG. 52 illustrates PTC Data Transfer in some embodiments.

FIG. 53 illustrates UE initiated PTC deactivation in some embodiments.

FIG. 54 illustrates UE initiated PTC re-activation in some embodiments.

FIG. 55 illustrates Network initiated PTC de-activation in some embodiments.

FIG. 56 illustrates Network initiated PTC re-activation in some embodiments.

FIG. 57 illustrates Implicit PTC deactivation in some embodiments.

FIG. 58 illustrates PDP Context Activation in some embodiments.

FIG. 59 illustrates Network Requested PDP Context Activation in some embodiments.

FIG. 60 illustrates UTRAN to GAN SRNS Relocation Preparation Phase in some embodiments.

FIG. 61 illustrates UTRAN to GAN SRNS Relocation Execution Phase in some embodiments.

FIG. 62 illustrates GAN to UTRAN SRNS Relocation Preparation Phase in some embodiments.

FIG. 63 illustrates GAN to UTRAN SRNS Relocation Execution Phase in some embodiments.

FIG. 64 illustrates the GAN architecture in support of the CS Domain control plane in some embodiments.

FIG. 65 illustrates the GAN protocol architecture in support of the CS domain user plane in some embodiments.

FIG. 66 illustrates the GAN architecture in support of the PS Domain Control Plane in some embodiments.

FIG. 67 illustrates the GAN architecture for the PS Domain User Plane in some embodiments.

FIG. 68 illustrates the GA-RC sublayer in the UE in some embodiments.

FIG. 69 illustrates successful (and unsuccessful) establishment of the GA-RRC Connection when initiated by the UE in some embodiments.

FIG. 70 illustrates successful establishment of the GA-RRC Connection when initiated by the network in some embodiments.

FIG. 71 shows release of the logical GA-RRC connection between the UE and the GANC in some embodiments.

FIG. 72 illustrates the message flow for security mode control in some embodiments.

FIG. 73 illustrates core network to UE NAS signaling of some embodiments.

FIG. 74 illustrates the UE to core network NAS signaling of some embodiments.

FIG. 75 illustrates mobile originated CS call procedure in some embodiments.

FIG. 76 illustrates an alternative procedure performed during a mobile originated CS call in some embodiments.

FIG. 77 illustrates mobile terminated CS call procedure in some embodiments.

FIG. 78 illustrates call clearing initiated by the UE in some embodiments.

FIG. 79 illustrates the CS Handover from GERAN to GAN procedure in some embodiments.

FIG. 80 illustrates an alternative procedure for CS handover from GERAN to GAN in some embodiments.

FIG. 81 illustrates the CS Handover from UTRAN to GAN procedure in some embodiments.

FIG. 82 illustrates an alternative procedure for CS handover from UTRAN to GAN using RRC protocol in some embodiments.

FIG. 83 illustrates the CS handover from GAN to GERAN procedure in some embodiments.

FIG. 84 illustrates the CS handover from GAN to UTRAN procedure in some embodiments.

FIG. 85 illustrates the Packet Transport Channel initial activation procedure of some embodiments.

FIG. 86 illustrates the transfer of GPRS user data packets via the GAN Packet Transport Channel in some embodiments.

FIG. 87 illustrates the scenario when the user equipment deactivates the Packet Transport Channel after the PTC Timer expires in some embodiments.

FIG. 88 illustrates the scenario when the user equipment initiates re-activation of the Packet Transport Channel in some embodiments.

FIG. 89 illustrates the scenario when the network initiates de-activation of the Packet Transport Channel in some embodiments.

FIG. 90 illustrates the scenario when the network initiates re-activation of the Packet Transport Channel in some embodiments.

FIG. 91 illustrates the successful user equipment initiated PDP Context Activation procedure in some embodiments.

FIG. 92 illustrates the successful Network-Requested PDP Context Activation procedure in some embodiments.

FIG. 93 illustrates the successful UE-initiated PDP Context Activation procedure in some embodiments.

FIG. 94 illustrates SRNS relocation procedure from UTRAN to GAN for a UE that is in PMM Connected state in some embodiments.

FIG. 95 conceptually illustrates a computer system with which some embodiments of the invention are implemented.

FIG. 96 illustrates the procedure for implicit PTC de-activation in some embodiments.

DETAILED DESCRIPTION OF THE INVENTION

In the following detailed description of the invention, numerous details, examples, and embodiments of the invention are set forth and described. However, it will be clear and apparent to one skilled in the art that the invention is not limited to the embodiments set forth and that the invention may be practiced without some of the specific details and examples discussed.

Throughout the following description, acronyms commonly used in the telecommunications industry for wireless services are utilized along with acronyms specific to the present invention. A table of acronyms used in this application is included in Section IX.

Several more detailed embodiments of the invention are described in sections below. Specifically, Section I describes the overall integrated communication system in which some embodiments are incorporated. The discussion in Section I is followed by a discussion of the functional entities of some embodiments in Section II. Next, Section III describes the control and user plane architecture of some embodiments. Section IV then describes the generic access network (GAN) security mechanism of some embodiments.

Next, Section V describes high level procedures such as discovery, registration, authentication, handover, etc. of some embodiments. Section VI then describes the configuration information of some embodiments. Next, identifiers used in GAN are presented in Section VII. An alternative embodiment that utilizes the same protocol for both voice and data services is disclosed in Section VIII. The discussion is followed by Section IX description of a computer system with which some embodiments of the invention are implemented. Finally, Section X lists the abbreviations used.

I. OVERALL SYSTEM

A. Integrated Communication Systems (ICS)

FIG. 1 illustrates an integrated communication system (ICS)architecture100 in accordance with some embodiments of the present invention.ICS architecture100 enables user equipment (UE)102 to access a voice anddata network165 via either alicensed air interface106 or anICS interface110 through which components of amobile core network165 are alternatively accessed. In some embodiments, a communication session includes voice services, data services, or both.

Themobile core network165 includes one or more Home Location Registers (HLRs)150 anddatabases145 for subscriber authentication and authorization. Once authorized, theUE102 may access the voice and data services of themobile core network165. In order to provide such services, themobile core network165 includes a mobile switching center (MSC)160 for providing access to the voice services. Data services are provided for through a Serving GPRS (General Packet Radio Service) Support Node (SGSN)155 in conjunction with a gateway such as the Gateway GPRS Support Node (GGSN)157.

TheSGSN155 is typically responsible for delivering data packets from and to theGGSN157 and the user equipment within the geographical service area of theSGSN155. Additionally, theSGSN155 may perform functionality such as mobility management, storing user profiles, and storing location information. However, the actual interface from themobile core network165 to various external data packet services networks (e.g., public Internet) is facilitated by theGGSN157. As the data packets originating from the user equipment typically are not structured in the format with which to access the external data networks, it is the role of theGGSN157 to act as the gateway into such packet services networks. In this manner, theGGSN157 provides addressing for data packets passing to and from theUE102 and the external packet services networks (not shown). Moreover, as the user equipment of a licensed wireless network traverses multiple service regions and thus multiple SGSNs, it is the role of theGGSN157 to provide a static gateway into the external data networks.

In the illustrated embodiment, components common to a UMTS Terrestrial Radio Access Network (UTRAN) basedcellular network185 are depicted that include multiple base stations referred to as Node Bs180 (of which only one is shown for simplicity) that facilitate wireless communication services forvarious user equipment102 via respective licensed radio links106 (e.g., radio links employing radio frequencies within a licensed bandwidth). However, one of ordinary skill in the art will recognize that in some embodiments, the licensed wireless network may include other licensed wireless networks such as the GSM/EDGE Radio Access Network (GERAN). An example of a system using A and Gb interfaces to access GERAN is shown inFIG. 3 below.

The licensedwireless channel106 may comprise any licensed wireless service having a defined UTRAN or GERAN interface protocol (e.g., Iu-cs and Iu-ps interfaces for UTRAN or A and Gb interfaces for GERAN) for a voice/data network. TheUTRAN185 typically includes at least oneNode B180 and a Radio Network Controller (RNC)175 for managing the set ofNode Bs180. Typically, themultiple Node Bs180 are configured in a cellular configuration (one per each cell) that covers a wide service area.

EachRNC175 communicates with components of thecore network165 through a standard radio network controller interface such as the Iu-cs and Iu-ps interfaces depicted inFIG. 1. For example, aRNC175 communicates withMSC160 via the UTRAN Iu-cs interface for circuit switched voice services. Additionally, theRNC175 communicates withSGSN155 via the UTRAN Iu-ps interface for packet data services throughGGSN157. Moreover, one of ordinary skill in the art will recognize that in some embodiments, other networks with other standard interfaces may apply. For example, theRNC175 in a GERAN network is replaced with a Base Station Controller (BSC) that communicates voice to theMSC160 via an A interface and the BSC communicates data to the SGSN via a Gb interface of the GERAN network.

In some embodiments of the ICS architecture, theuser equipment102 use the services of the mobile core network (CN)165 via a second communication network facilitated by theICS access interface110 and a Generic Access Network Controller (GANC)120 (also referred to as a Universal Network Controller or UNC).

In some embodiments, the voice and data services over theICS access interface110 are facilitated via anaccess point114 communicatively coupled to abroadband IP network116. In some embodiments, theaccess point114 is a generic wireless access point that connects theuser equipment102 to the ICS network through anunlicensed wireless network118 created by theaccess point114.

The signaling from theUE102 is passed over theICS access interface110 to theGANC120. After theGANC120 performs authentication and authorization of the subscriber, theGANC120 communicates with components of themobile core network165 using a radio network controller interface that is the same or similar to the radio network controller interface of the UTRAN described above, and includes a UTRAN Iu-cs interface for circuit switched voice services and a UTRAN Iu-ps interface for packet data services (e.g., GPRS). In this manner, theGANC120 uses the same or similar interface to the mobile core network as a UTRAN Radio Network Subsystem (e.g., theNode B180 and RNC175).

In some embodiments, theGANC120 communicates with other system components of the ICS system through one or more of several other interfaces, which are (1) “Up”, (2) “Wm”, (3) “D′/Gr′”, (4) “Gn′”, and (5) “S1”. The “Up” interface is the interface between theUE102 and theGANC120. The “Wm” interface is a standardized interface between theGANC120 and an Authorization, Authentication, and Accounting (AAA)Server170 for authentication and authorization of theUE102 into the ICS. The “D′/Gr′” interface is the standard interface between theAAA server170 and theHLR160. Optionally, some embodiments use the “Gn′” interface which is a modified interface for direct communications with the data services gateway (e.g., GGSN) of the core licensed network. Some embodiments optionally include the “S1” interface. In these embodiments, the “S1” interface provides an authorization and authentication interface from theGANC120 to anAAA140 server. In some embodiments, theAAA server140 that supports the S1 interface and theAAA server170 that supports Wm interface may be the same. More details of the S1 interface are described in U.S. application Ser. No. 11/349,025, entitled “Service Access Control Interface for an Unlicensed Wireless Communication System”, filed Feb. 6, 2006.

In some embodiments, theUE102 must register with theGANC120 prior to accessing ICS services. Registration information of some embodiments includes a subscriber's International Mobile Subscriber Identity (IMSI), a Media Access Control (MAC) address, and a Service Set Identifier (SSID) of the serving access point as well as the cell identity from the GSM or UTRAN cell upon which theUE102 is already camped. In some embodiments, theGANC120 may pass this information to theAAA server140 to authenticate the subscriber and determine the services (e.g., voice and data) available to the subscriber. If approved by theAAA140 for access, theGANC120 will permit theUE102 to access voice and data services of the ICS system.

These voice and data services are seamlessly provided by the ICS to theUE102 through the various interfaces described above. In some embodiments, when data services are requested by theUE102, the ICS uses the optional Gn′ interface for directly communicating with aGGSN157. The Gn′ interface allows theGANC120 to avoid the overhead and latency associated with communicating with theSGSN155 over the Iu-ps interface of the UTRAN or the Gb interface of the GSM core networks prior to reaching theGGSN157.

In some other embodiments, theaccess point114 is a Femtocell access point (FAP). The FAP facilitates short-range licensedwireless communication sessions118 that operate independent of the licensedcommunication session106. In case of the Femtocell, theuser equipment102 connects to the ICS network through the short-range licensedwireless network118 created by theFAP114. Signals from the FAP are then transmitted over thebroadband IP network116.

B. Applications of ICS

An ICS provides scalable and secure interfaces into the core service network of mobile communication systems.FIG. 2 illustrates several applications of an ICS in some embodiments. As shown, homes, offices, hot spots, hotels, and other public andprivate places205 are connected to one or more network controllers210 (such as theGANC120 shown inFIG. 1) through theInternet215. The network controllers in turn connect to the mobile core network220 (such as thecore network165 shown inFIG. 1).

FIG. 2 also shows several user equipments. These user equipments are just examples of user equipments that can be used for each application. Although in most examples only one of each type of user equipments is shown, one of ordinary skill in the art would realize that other type of user equipments can be used in these examples without deviating from the teachings of the invention. Also, although only of each type of access points, user equipment, or network controllers are shown, many such access points, user equipments, or network controllers may be employed inFIG. 2. For instance, an access point may be connected to several user equipment, a network controller may be connected to several access points, and several network controllers may be connected to the core network. The following sub-sections provide several examples of services that can be provided by an ICS.

1. Wi-Fi

A Wi-Fi access point230 enables a dual-mode cellular/Wi-Fi UEs260-265 to receive high-performance, low-cost mobile services when in range of a home, office, or public Wi-Fi network. With dual-mode UEs, subscribers can roam and handover between licensed wireless communication system and Wi-Fi access and receive a consistent set of services as they transition between networks.

2. Femtocells

A Femtocell enables user equipments, such as standardmobile stations270 and wireless enabledcomputers275 shown, to receive low cost services using a short-range licensed wireless communication sessions through aFAP235.

3. Terminal Adaptors

Terminal adaptors

240 allow incorporating fixed-terminal devices such astelephones245,Faxes250, and other equipments that are not wireless enabled within the ICS. As long as the subscriber is concerned, the service behaves as a standard analog fixed telephone line. The service is delivered in a manner similar to other fixed line VoIP services, where a UE is connected to the subscriber's existing broadband (e.g., Internet) service.

4. WiMAX

Some licensed wireless communication system operators are investigating deployment of WiMAX networks in parallel with their existing cellular networks. A dual mode cellular/WiMAX UE290 enables a subscriber to seamlessly transition between a cellular network and such a WiMAX network.

5. SoftMobiles

Connectinglaptops280 to broadband access at hotels and Wi-Fi hot spots has become popular, particularly for international business travelers. In addition, many travelers are beginning to utilize their laptops and broadband connections for the purpose of voice communications. Rather than using mobile phones to make calls and pay significant roaming fees, they utilize SoftMobiles (or SoftPhones) and VoIP services when making long distance calls.

To use a SoftMobile service, a subscriber would place aUSB memory stick285 with an embedded SIM into a USB port of theirlaptop280. A SoftMobile client would automatically launch and connect over IP to the mobile service provider. From that point on, the subscriber would be able to make and receive mobile calls as if she was in her home calling area.

Several examples of Integrated Communication Systems (ICS) are given in the following sub-sections. A person of ordinary skill in the art would realize that the teachings in these examples can be readily combined. For instance, an ICS can be an IP based system and have an A/Gb interface towards the core network while another ICS can have a similar IP based system with an Iu interface towards the core network.

C. Integrated Systems with A/Gb and/or Iu Interfaces towards the Core Network

FIG. 3 illustrates the A/Gb-mode Generic Access Network (GAN) functional architecture of some embodiments. The GAN includes one or more Generic Access Network Controllers (GANC)310 and one or more genericIP access networks315. One or more UEs305 (one is shown for simplicity) can connect to aGANC310 through a genericIP access network315. TheGANC310 has the capability to appear to thecore network325 as a GSM/EDGE Radio Access Network (GERAN) Base Station Controller (BSC). TheGANC310 includes a Security Gateway (SEGW)320 that terminates secure remote access tunnels from theUE305, providing mutual authentication, encryption and data integrity for signaling, voice and data traffic.

The genericIP access network315 provides connectivity between theUE305 and theGANC310. The IP transport connection extends from theGANC310 to theUE305. A single interface, the Up interface, is defined between theGANC310 and theUE305.

The GAN co-exists with the GERAN and maintains the interconnections with the Core Network (CN)325 via the standardized interfaces defined for GERAN. These standardized interfaces include the A interface to Mobile Switching Center (MSC)330 for circuit switched services, Gb interface to Serving GPRS Support Node (SGSN)335 for packet switched services, Lb interface to Serving Mobile Location Center (SMLC)350 for supporting location services, and an interface to Cell Broadcast Center (CBC)355 for supporting cell broadcast services. The transaction control (e.g. Connection Management, CC, and Session Management, SM) and user services are provided by the core network (e.g. MSC/VLR and the SGSN/GGSN).

As shown, theSEGW320 is connected to aAAA server340 over the Wm interface. TheAAA server340 is used to authenticate theUE305 when it sets up a secure tunnel. Some embodiments require only a subset of the Wm functionalities for the GAN application. In these embodiments, as a minimum the GANC-SEGW shall support the Wm authentication procedures.

FIG. 4 illustrates the Iu-mode Generic Access Network (GAN) functional architecture of some embodiments. The GAN includes one or more Generic Access Network Controllers (GANC)410 and one or more genericIP access networks415. One or more UEs405 (one is shown for simplicity) can be connected to aGANC410 through a genericIP access network415. In comparison with theGANC310, theGANC410 has the capability to appear to thecore network425 as a UMTS Terrestrial Radio Access Network (UTRAN) Radio Network Controller (RNC). In some embodiments, the GANC has the expanded capability of supporting both the Iu and A/Gb interfaces to concurrently support both Iu-mode and A/Gb-mode UEs. Similar to theGANC310, theGANC410 includes a Security Gateway (SEGW)420 that terminates secure remote access tunnels from theUE405, providing mutual authentication, encryption and data integrity for signaling, voice and data traffic.

The genericIP access network415 provides connectivity between theUE405 and theGANC410. The IP transport connection extends from theGANC410 to theUE405. A single interface, the Up interface, is defined between theGANC410 and theUE405. Functionality is added to this interface, over the UP interface shown inFIG. 3 to support the Iu-mode GAN service.

The GAN co-exists with the UTRAN and maintains the interconnections with the Core Network (CN)425 and via the standardized interfaces defined for UTRAN. These standardized interfaces include the Iu-cs interface to Mobile Switching Center (MSC)430 for circuit switched services, Iu-ps interface to Serving GPRS Support Node (SGSN)435 for packet switched services, Iu-pc interface to Serving Mobile Location Center (SMLC)450 for supporting location services, and Iu-bc interface to Cell Broadcast Center (CBC)455 for supporting cell broadcast services. The transaction control (e.g. Connection Management, CC, and Session Management, SM) and user services are provided by the core network (e.g. MSC/VLR and the SGSN/GGSN).

As shown, theSEGW420 is connected to aAAA server440 over the Wm interface. TheAAA server440 is used to authenticate theUE405 when it sets up a secure tunnel. Some embodiments require only a subset of the Wm functionalities for the Iu mode GAN application. In these embodiments, as a minimum the GANC-SEGW shall support the Wm authentication procedures.

D. ATM and IP Based Architectures

In some embodiments, the system uses Asynchronous Transfer Mode (ATM) based Iu (Iu-cs and Iu-ps) interfaces towards the CN. In some embodiments, the system architecture can also support an IP based Iu (Iu-cs and Iu-ps) interface towards the CN. The following two sub-sections describe examples of these architectures for Femtocell.

A person of ordinary skill in the art would realize that the same examples can be readily applied to other types of ICS. For instance, these examples can be used when the ICS access interface110 (shown inFIG. 1) uses unlicensed frequencies (instead of Femtocell's licensed frequencies), theaccess point114 is a generic WiFi access point (instead of a FAP), etc. Also, a person of ordinary skill in the art would realize that the same examples can be readily implemented using A/Gb interfaces (described above) instead of Iu interfaces.

FIG. 5 illustrates the basic elements of a Femtocell system architecture with Asynchronous Transfer Mode (ATM) based Iu (Iu-cs and Iu-ps) interfaces towards the CN in some embodiments. These elements include the user equipment (UE)505, theFAP510, and the Generic Access Network Controller (GANC)515, and the Access Point Management SYSTEM (AMS)570.

For simplicity, only one UE and one FAP are shown. However, each GANC can support multiple FAPs and each FAP in turn can support multiple UEs. As shown, theGANC515 includes an IP Network Controller (INC)525, a GANC Security Gateway (SeGW)530, aGANC Signaling Gateway535, a GANC Media Gateway (MGW)540, an ATM Gateway (545). Elements of the Femtocell are described further below.

FIG. 6 illustrates the basic elements of a Femtocell system architecture with an IP based Iu (Iu-cs and Iu-ps) interface towards the CN in some embodiments. For simplicity, only one UE and one FAP are shown. However, each GANC can support multiple FAPs and each FAP in turn can support multiple UEs. This option eliminates the need for theGANC Signaling gateway535 and also theATM gateway545. Optionally for IP based Iu interface, theGANC Media Gateway540 can also be eliminated if the R4 MGW605 in the CN can support termination of voice data i.e. RTP frames as defined in “IETF RFC 3267—Real-Time Transport Protocol (RTP) Payload Format and File Storage Format for the Adaptive Multi-Rate (AMR) and Adaptive Multi-Rate Wideband (AMR-WB) Audio Codecs”, “RFC 3267”.

Also shown inFIGS. 5 and 6 are components of the licensed wireless communication systems. These components are

3G MSC

550,3G SGSN555, and other Core Network System (shown together)565. The3G MSC550 provides a standard Iu-cs interface towards the GANC. Another alternative for the MSC is shown inFIG. 6. As shown, theMSC650 is split up into a MSS (MSC Server)675 for Iu-cs based signaling andMGW680 for the bearer path.R4 MSC650 is arelease 4 version of a 3G MSC with a different architecture i.e. R4 MSC is split into MSS for control traffic and a MGW for handling the bearer. A similar MSC can be used for the ATM architecture ofFIG. 5. Both architectures shown inFIGS. 5 and 6 are also adaptable to use any future versions of the MSC.

The3G SGSN555 provides packet services (PS) via the standard Iu-ps interface. The SGSN connects to theINC525 for signaling and to theSeGW530 for PS data. TheAAA server560 communicates with theSeGW530 and supports the EAP-AKA and EAP-SIM procedures used in IKEv2 over the Wm interface and includes a MAP interface to the HLR/AuC. In some embodiments, this system also supports the enhanced service access control functions over the S1 interface.

II. FUNCTIONAL ENTITIES

A. User Equipment

TheUE405 contains the functions that are required to access the Iu-mode GAN. In some embodiments, the UE additionally contains that are required to access the A/Gb-mode GAN. In some embodiments, the User Equipment (UE)305 is a dual mode (e.g., GSM and unlicensed radios) handset device with capability to switch between the two modes. The user equipment can support either Bluetooth® or IEEE 802.11 protocols. In some embodiments, the UE supports an IP interface to the access point. In these embodiments, the IP connection from the GANC extends all the way to the UE. In some other embodiments, the User Equipment (UE)305 is a standard 3G handset device operating over licensed spectrum of the provider.

In some embodiments, the user equipment includes a cellular telephone, smart phone, personal digital assistant, or computer equipped with a subscriber identity mobile (SIM) card for communicating over the licensed or unlicensed wireless networks. Moreover, in some embodiments the computer equipped with the SIM card communicates through a wired communication network.

Alternatively, in some embodiments the user equipment includes a fixed wireless device providing a set of terminal adapter functions for connecting Integrated Services Digital Network (ISDN), Session Initiation Protocol (SIP), or Plain Old Telephone Service (POTS) terminals to the ICS. Application of the present invention to this type of device enables the wireless service provider to offer the so-called landline replacement service to users, even for user locations not sufficiently covered by the licensed wireless network. Moreover, some embodiments of the terminal adapters are fixed wired devices for connecting ISDN, SIP, or POTS terminals to a different communication network (e.g., IP network) though alternate embodiments of the terminal adapters provide wireless equivalent functionality for connecting through unlicensed or licensed wireless networks.

B. Generic Access Network Controller (GANC)

Thecore network425 interacts with theGANC410 as though it was an RNC. The genericIP access network415 provides connectivity between theGANC410 and theUE405. TheGANC410 entity inter-works between the Iu interfaces and a generic IP access network, using the control plane and user plane functionalities. The control plane functionality is utilized for call control signaling and the user plane functionality is utilized for information transfer (e.g., voice or data). In some embodiments, the GANC has the extended capability to also inter-work with GERAN A/Gb interfaces.

Some embodiments of the above mentioned devices, such as the user equipment, FAP, or GANC, include electronic components, such as microprocessors and memory (not shown), that store computer program instructions for executing wireless protocols for managing voice and data services in a machine-readable or computer-readable medium as further described below in the section labeled “Computer System”. Examples of machine-readable media or computer-readable media include, but are not limited to magnetic media such as hard disks, memory modules, magnetic tape, optical media such as CD-ROMS and holographic devices, magneto-optical media such as optical disks, and hardware devices that are specially configured to store and execute program code, such as application specific integrated circuits (ASICs), programmable logic devices (PLDs), ROM, and RAM devices. Examples of computer programs or computer code include machine code, such as produced by a compiler, and files containing higher-level code that are executed by a computer, an electronic component, or a microprocessor using an interpreter.

III. CONTROL AND USER PLANE ARCHITECTURE

In some embodiments, the Iu interface includes support for both Asynchronous Transfer Mode (ATM) and IP-based signaling and user data transport mechanisms. The following sections describe the control and user plane architectures for the Circuit Switched (CS) domain and Packet Switched (PS) domain of some embodiments.

A. Circuit Switched (CS) Domain

1. CS Domain—Control Plane

FIG. 7 illustrates the GAN architecture in support of the CS Domain control plane in some embodiments. The figure shows different protocol layers for theUE705,Generic IP Network710,GANC715, andMSC720.FIG. 7 also shows the two interfaces Up725 and Iu-cs730. The main features of the GAN CS domain control plane architecture are as follows. Theunderlying Access Layers735 andTransport IP layer740 provide the generic IP connectivity between theUE705 and theGANC715. TheIPSec layer745 provides encryption and data integrity between theUE705 andGANC715. TheRemote IP layer750 is the ‘inner’ IP layer for IPSec tunnel mode and is used by theUE705 to be addressed by theGANC715. TheRemote IP layer750 is configured during the IPSec connection establishment.

In some embodiments, a single TCP connection is used to provide reliable transport for both the GA-RC and GA-CSR signaling between theUE705 andGANC715. The TCP connection is managed by GA-RC and is transported using the Remote IP layer. Non-Access Stratum (NAS) protocols, such asMM760 and above, are carried transparently between theUE705 andMSC720. The Generic Access Resource Control (GA-RC) protocol manages the Up session, including the GAN discovery and registration procedures. The GA-RC protocol (described in clause 8.1.4 of “Generic access to the A/Gb interface;Stage 2”, 3GPP TS 43.318 standard) is extended to include support for the selection of either A/Gb mode or Iu mode GAN.

The Generic Access Circuit Switched Resource (GA-CSR) protocol supports UMTS-specific requirements as well as GERAN-specific requirements. TheGANC715 terminates the GA-CSR protocol and inter-works it to theRANAP755 protocol over the Iu-cs730 interface. In some embodiments, the Iu-cssignaling transport layers765 are per “UTRAN Iu interface signalling transport”, 3GPP TS 25.412 standard, hereinafter “3GPP TS 25.412”.

a) Alternative Architectures for CS Domain—Control Plane

The embodiment shown inFIG. 7 is just one alternative for implementing the CS domain control plane architecture in which aUE705 and aGeneric IP Network710 are used to connect a subscriber using the UE to theMSC720 through theGANC715. A person of ordinary skill in the art would realize that the teachings of the invention can be applied for other user equipment and access points (such as the ones described inFIG. 2).

For instance,FIG. 8 illustrates the CS domain control plane architecture of some embodiments. As shown, the GANC and MSC inFIG. 8 are similar to the GANC and MSC shown inFIG. 7. InFIG. 8, the local node in which the subscriber is located is represented as a black box (referred to as Local Node805). Different embodiments utilize different equipments in order to connect a subscriber located in theLocal Node805 with theMSC720 through theGANC715. For instance, in the embodiment shown inFIG. 7, aUE705 and a Generic IP Network are used710.FIG. 9 illustrates another embodiment in which aUE905, a Femtocell access point (FAP)910, and aGeneric IP Network915 are used to connect theLocal Node805 with theMSC720 through theGANC715.

As shown, the protocol layers of the GANC880-885 are communicatively coupled (shown with arrows845-850 respectively) with their corresponding layers in theGeneric IP Network915. Similarly, the GANC layers855-875 are communicatively coupled (shown with arrows820-840 respectively) with their corresponding layers in theFAP910. Also theMM layer890 and CC/CS/SMS layers895 of theMSC720 are transparently connected (shown with arrows810-815 respectively) to their corresponding layers in theUE905. Using this technique, the FAP similar to theFAP235 shown inFIG. 2 can be utilized to connect a UE (such as UEs270-275) to thewireless core network220 through anetwork controller210. A person of ordinary skill in the art would be able to apply the technique shown inFIGS. 8 and 9 to communicatively couple any user equipment, access points, terminal adaptors, SoftMobiles, etc. (such as the ones shown inFIG. 2) to an integrated communication system (ICS) that uses a multi-layer CS domain control architecture as shown inFIG. 7.

b) CS Domain—Control Plane—UE Architecture

FIG. 10 illustrates the UE architecture for the CS domain control plane. As shown, the architecture includes support for GERAN, UTRAN, and both A/Gb mode GAN and Iu mode GAN. The main features of the UE CS Domain Control Plane architecture shown inFIG. 10 are as follows. The GERAN RR-SAP interface1015 to the GSM-MM layer1005 is preserved identically for both GERAN and A/Gb-mode GAN access. Likewise, the UTRAN RR-SAP interface1020 to the GSM-MM layer1005 is preserved identically for both UTRAN and Iu-mode GAN access. An access mode switch1010 is provided to switch between GERAN/UTRAN, A/GB-mode GAN and Iu-mode GAN modes. GA-CSR/GA-RC1025 peers directly with theUTRAN RRC1030 andGERAN RRC1035 layers to provide coordination for roving and handover. As shown inFIG. 10, GA-CSR/GA-RC1025,UTRAN RRC1030, andGERAN RRC1035 interface through a set of service access interfaces (SAPs)1040.

2. CS Domain—User Plane

FIG. 11 illustrates the GAN protocol architecture in support of the CS domain user plane in some embodiments. The figure shows different protocol layers for theUE1105,Generic IP Network1110,GANC1115, andMSC1120.FIG. 11 also shows the two interfaces Up1125 and Iu-cs1130. The main features of the GAN CS domain user plane architecture are as follows. Theunderlying Access Layers1135 andTransport IP layer1140 provide the generic connectivity between theUE1105 and theGANC1115. TheIPSec layer1145 provides encryption and data integrity. The CS user plane data transport over theUp interface1125 is the same as the CS user plane for A/Gb-mode GAN (i.e., using the Real Time Protocol, RTP, perIETF RFC 3267. TheGANC1115 interworks the CS domain user plane between RTP/UDP and the Iu User Plane (Iu-UP) protocol on the Iu-cs interface1130. In some embodiments, the Iu-csData transport layers1165 are per 3GPP TS 25.414 standard.

A person of ordinary skill in the art would realize that other user equipments, access point, terminal adaptor, SoftMobiles, etc. can be connected to the core network through a GANC. For instance,FIG. 12 illustrates the CS domain, user plane protocol architecture of aUE1205, a Femtocell access point (FAP)1210, andGeneric IP Network1215. Using the technique described in conjunction withFIGS. 8 and 9, a person of ordinary skill in the art would be able to replace theUE1105 andGeneric IP Network1110 shown inFIG. 11 with theUE1205,FAP1210, andGeneric IP Network1215 to connect theFemtocell UE1205 to the core network through the GANC. Similarly, other types of UE, access points, terminal adaptors, SoftMobiles, etc. can be connected to the core network through the GANC.

b) CS Domain—User Plane—UE Architecture

FIG. 13 illustrates the UE architecture for the CS domain user plane in some embodiments. As shown, the architecture includes support for both A/Gb mode andIu mode GAN1305, as well asGERAN1310, andUTRAN1315. TheRFC 3267AMR Processing layer1320 is utilized for connecting the GAN RTP/UDP/IP layers1325 to the AMRAudio Processing layer1330 through the CS User PlaneRouting Service layer1335, which routes CS user plane data to and from the selected access network; i.e., GERAN, UTRAN, or GAN. TheRFC 3267AMR Processing layer1320 is not used when connecting to the CSData Processing layer1340; i.e., in the case of circuit switched data, as opposed to circuit switched voice.

B. Packet Switched (PS) Domain

1. PS Domain—Control Plane

FIG. 14 illustrates the GAN architecture in support of the PS Domain Control plane. The figure shows different protocol layers for theUE1405,Generic IP Network1410,GANC1415,SGSN1420.FIG. 14 also shows the two interfaces Up1425 and Iu-ps1430. The main features of the GAN PS domain control plane architecture shown inFIG. 14 are as follows. Theunderlying Access Layers1435 andTransport IP layer1440 provide the generic connectivity between theUE1405 and theGANC1415. TheIPSec layer1445 provides encryption and data integrity.TCP1450 provides reliable transport for the GA-PSR betweenUE1405 andGANC1415. The GA-RC manages the IP connection, including the GAN registration procedures. The Generic Access Packet Switched Resource (GA-PSR) protocol supports UMTS-specific requirements.

TheGANC1415 terminates the GA-PSR protocol and inter-works it to theRANAP protocol1455 over the Iu-ps interface1430.NAS protocols1460, such as for GMM, SM and SMS, are carried transparently between theUE1405 andSGSN1420. In some embodiments, the Iu-pssignaling transport layers1465 are per 3GPP TS 25.412.

A person of ordinary skill in the art would realize that other user equipments, access point, terminal adaptor, SoftMobiles, etc. can be connected to the core network through a GANC. For instance,FIG. 15 illustrates the PS domain, control plane protocol architecture of aUE1505, a Femtocell access point (FAP)1510, andGeneric IP Network1515. Using the technique described in conjunction withFIGS. 8 and 9, a person of ordinary skill in the art would be able to replace theUE1405 andGeneric IP Network1410 shown inFIG. 11 with theUE1505,FAP1510, andGeneric IP Network1515 to connect theFemtocell UE1505 to the core network through the GANC. Similarly, other types of UE, access points, terminal adaptors, SoftMobiles, etc. can be connected to the core network through the GANC.

c) PS Domain—Control Plane—UE Architecture

FIG. 16 illustrates the UE architecture for the PS domain control plane in some embodiments. As shown, the architecture includes support for both A/Gb mode and Iu mode GAN, as well as GERAN and UTRAN. The main features of the UE PS Domain Control Plane architecture shown inFIG. 16 are as follows. The GERAN GRR-SAP interface1615 and GERAN GMMRR-SAP interface1617 to theGMM layer1605 is preserved identically for both GERAN and A/Gb-mode GAN access. Likewise, the UTRAN RABMAS-SAP interface1620 and UTRAN GMMAS-SAP interface1622 to theGMM layer1605 is preserved identically for both UTRAN and Iu-mode GAN access. Anaccess mode switch1610 is provided to switch between GERAN/UTRAN, A/GB-mode GAN and Iu-mode GAN modes. GA-PSR/GA-RC1625 peers directly with theUTRAN RRC1630 andGERAN RRC1635 layers to provide coordination for roving and handover. As shown inFIG. 16, GA-PSR/GA-RC1625,UTRAN RRC1630, andGERAN RRC1635 interface through a set of service access interfaces (SAPs)1640.

2. PS Domain—User Plane

FIG. 17 illustrates the GAN architecture for the PS Domain User Plane in some embodiments. The figure shows different protocol layers for theUE1705,Generic IP Network1710,GANC1715,SGSN1720.FIG. 17 also shows the two interfaces Up1725 and Iu-ps1730. The main features of the GAN PS domain user plane architecture shown inFIG. 17 are as follows. Theunderlying Access Layers1735 andTransport IP layer1740 provide the generic connectivity between theUE1705 and theGANC1715. TheIPSec layer1745 provides encryption and data integrity.

GA-PSR is extended to include support for the GTP-U G-PDU message format to transport PS User Data (e.g., IP packets), rather than LLC PDUs as in A/Gb mode GAN. As shown inFIG. 17, user data in GTP-U G-PDU messages may be carried transparently between theUE1705 and core network through the SGSN to the GGSN. In some embodiments, the Iu-ps data transportlower layers1765 are per 3GPP TS 25.414 standard.

FIG. 18 illustrates an alternative GAN PS domain user plane configuration which is supported by the Up interface procedures of some embodiments. In this configuration, theGANC1815 terminates the Up interface GTP-U tunnel with theUE1805 and also terminates the separate Iu-ps GTP-U tunnel to theSGSN1820. TheGANC1815 relays the PS user data between the Up interface GTP-U tunnel and the associated Iu-ps interface GTP-U tunnel to allow the PS user data to flow between the UE and the SGSN.

This configuration minimizes the number of active GTP-U “paths” presented to the core network; i.e., the SGSN may be limited in the number of RNCs with which it can concurrently exchange PS user data (e.g., today, there can be no more than 4096 RNCs in a given PLMN). It may not be able to support-without a software upgrade, for example—concurrent communication with hundreds of thousands of UEs as would be required if the GTP-U tunnels were from UE to SGSN. Terminating the Iu-ps GTP-U tunnels on the GANC avoids this potential SGSN limitation. In some embodiments, the Iu-ps data transportlower layers1865 are per 3GPP TS 25.414 standard.

A person of ordinary skill in the art would realize that other user equipments, access point, terminal adaptor, SoftMobiles, etc. can be connected to the core network through a GANC. For instance,FIG. 19 illustrates the PS domain, user plane protocol architecture of aUE1905, a Femtocell access point (FAP)1910, andGeneric IP Network1915. Using the technique described in conjunction withFIGS. 8 and 9, a person of ordinary skill in the art would be able to replace theUE1805 andGeneric IP Network1810 shown inFIG. 11 with theUE1905,FAP1910, andGeneric IP Network1915 to connect theFemtocell UE1905 to the core network through the GANC. Similarly, other types of UE, access points, terminal adaptors, SoftMobiles, etc. can be connected to the core network through the GANC.

a) PS Domain—User Plane—UE Architecture

FIG. 20 illustrates the UE architecture for the PS domain user plane in some embodiments. As shown, the architecture includes support for both A/Gb mode andIu mode GAN2005, as well asGERAN2010, andUTRAN2015. Anaccess mode switch2020 is provided to switch between GERAN/UTRAN, A/GB-mode GAN and Iu-mode GAN modes.

C. GA-RC (Generic Access Resource Control)

The GA-RC protocol provides a resource management layer, with the following functions. Discovery and registration with GANC, registration update with GANC, application level keep-alive with GANC; and support for identification of the AP being used for GAN access.

1. States of the GA-RC Sub-Layer

FIG. 21 illustrates the state diagram for generic access in the UE in some embodiments. As shown, the GA-RC sub-layer in the UE can be in one of two states: GA-RC-DEREGISTERED2105 or GA-RC-REGISTERED2110. The following outcomes are possible when switching (shown by arrow2112) the serving RR to Iu-mode GAN: (1) Transition to GA-CSR-IDLE2115 and GA-PSR-IDLE2120 (i.e., if the UE is idle during the transition), (2) Transition to GA-CSR-CONNECTED2125 and GA-PSR-IDLE2130 (i.e., due to CS handover or relocation), (3) Transition to GA-CSR-IDLE2115 and GA-PSR-CONNECTED2130 (i.e., due to PS handover or relocation), (4) Transition to GA-CSR-CONNECTED2125 and GA-PSR-CONNECTED2130 (i.e., due to dual transfer mode handover or CS+PS relocation). The switch of the serving RR from GAN to GERAN/UTRAN RRC (shown by arrow2135) may occur when the UE is in any combination of the GA-CSR and GA-PSR states.

In the GA-RC-DEREGISTERED state2105, the UE may be in a GAN coverage area; however, the UE has not registered successfully with the GANC. The UE may initiate the GAN Registration procedure when in the GA-RC-DEREGISTERED state2105. The UE returns to GA-RC-DEREGISTERED state2105 on loss of TCP or IPSec connection or on execution of the GAN De-registration procedure.

In the GA-RC-REGISTEREDstate2110, the UE is registered with the Serving GANC. The UE has an IPSec tunnel and a TCP connection established to the Serving GANC through which the UE may exchange GA-RC, GA-CSR, and GA-PSR signaling messages with the GANC.

While the UE remains in the GA-RC-REGISTEREDstate2110 it performs application level keep-alive with the GANC. In the GA-RC-REGISTEREDstate2110, the UE may be in either UTRAN/GERAN mode or GAN mode. The UE may either (1) be camped on GERAN or UTRAN and idle, (2) be active in GERAN or UTRAN (e.g., a GSM RR or a UTRAN RRC connection may be established), (3) have “roved in” to GAN mode, or (4) have recently “roved out” of GAN mode (e.g., due to handover from GAN).

D. GA-CSR (Generic Access Circuit Switched Resources)

The GA-CSR protocol provides a circuit switched services resource management layer which supports the following functions: (1) setup of transport channels for CS traffic between the UE and GANC, (2) CS handover support between UTRAN/GERAN and GAN, (3) direct transfer of NAS messages between the UE and the core network, and (4) other functions such as CS paging and security configuration.

1. States of the GA-CSR Sub-Layer

The GA-CSR sub-layer in the UE can be in two states, GA-CSR-IDLE or GA-CSR-CONNECTED as illustrated inFIG. 21. The UE enters the GA-CSR-IDLE state2115 when the UE switches the serving RR entity to GAN. This switch may occur only when the GA-RC is in the GA-RC-REGISTEREDstate2110.

The UE moves from the GA-CSR-IDLE state2115 to the GA-CSR-CONNECTEDstate2125 when the GA-CSR connection is established and returns to GA-CSR-IDLE state2115 when the GA-CSR connection is released. Upon GA-CSR connection release, an indication that no dedicated CS resources exist is passed to the upper layers. The UE may also enter the GA-CSR-CONNECTEDstate2125 while in the GA-RC-REGISTEREDstate2110 in GERAN/UTRAN mode when Handover to GAN is being performed. In the same way, the UE enters the GA-RC-REGISTEREDstate2110 in GERAN/UTRAN mode from the GA-CSR-CONNECTEDstate2125 when Handover from GAN is successfully executed.

E. GA-PSR (Generic Access Packet Switched Resources)

The GA-PSR protocol provides a packet switched services resource management layer which supports the following functions: (1) setup of transport channels for PS traffic between the UE and network, (2) PS relocation/handover support between UTRAN/GERAN and GAN, (3) direct transfer of NAS messages between the UE and the PS core network, (4) transfer of GPRS user plane data, and (5) other functions such as PS paging and security configuration.

1. States of the GA-PSR Sub-Layer

The GA-PSR sub-layer in the UE can be in two states, GA-PSR-IDLE or GA-PSR-CONNECTED as illustrated inFIG. 21. The UE enters the GA-PSR-IDLE state2120 when the UE switches the serving RR entity to GAN. This switch may occur only when the GA-RC is in the GA-RC-REGISTEREDstate2110. The UE moves from the GA-PSR-IDLE state2120 to the GA-PSR-CONNECTEDstate2130 when the GA-PSR connection is established and returns to GA-PSR-IDLE state2120 when the GA-PSR connection is released. Upon GA-PSR connection release, an indication that no dedicated resources exist is passed to the upper layers.

The UE may also enter the GA-PSR-CONNECTEDstate2130 while in the GA-RC-REGISTEREDstate2110 in GERAN/UTRAN mode when Handover to GAN is being performed. In the same way, the UE enters the GA-RC-REGISTEREDstate2110 in GERAN/UTRAN mode from the GA-PSR-CONNECTEDstate2130 when Handover from GAN is successfully executed. The GA-PSR Packet Transport Channel (GA-PSR PTC) provides the association between the UE and GANC for the transport of GPRS user data over the Up interface. It is described in PS NAS Signaling Procedures in sub-section V.P, below.

IV. GAN SECURITY MECHANISMS

GAN supports security mechanisms at different levels and interfaces as depicted inFIG. 22. Thesecurity mechanisms2205 over the Up interface protect control plane and user plane traffic flows between theUE2210 and theGANC2215 from unauthorized use, data manipulation and eavesdropping; i.e., authentication, encryption and data integrity mechanisms are supported.

Network access security

2220 includes the mechanisms defined in “3G Security; Security Architecture”, 3GPP TS 33.102 standard. Mutual authentication of the subscriber and the core network (CN)2225 occurs between the MSC/VLR or SGSN and the UE and is transparent to the GANC. However, there is a cryptographic binding between the UE-CN authentication and the UE-GANC authentication to prevent man-in-the-middle attacks.

Additional applicationlevel security mechanisms2230 may be employed in the PS domain to secure the end-to-end communication between theUE2210 and the application server2235. For example, in some embodiments theUE2210 may run the HTTP protocol over an SSL session for secure web access.

All control plane and user plane traffic sent between theUE2210 and theGANC2215 over the Up interface is protected by an IPSec tunnel between theUE2210 and GANC-SEGW, that provides mutual authentication (using USIM credentials), encryption and data integrity using the same mechanisms as specified in “3G security; Wireless Local Area Network (WLAN) interworking security”, 3GPP TS 33.234.

As described above (in relation toFIGS. 9, 12,15, and19), some embodiments utilize a Femtocell access point (FAP) to communicatively couple a user equipment UE to the GANC via a Generic IP Network. As shown inFIG. 9, the FAP architecture for the CS control plane has anIPSec layer920. Similarly the FAP architectures for the CS user plane, PS control plane, and PS user plane architectures also include IPSec (or IPSec ESP) layers (1220,1520, and1920 respectively). As shown inFIGS. 9, 12,15, and19, these IPSec layers are over the transport IP layer and Remote IP layers of the GANC and are communicatively coupled to their corresponding GANC IPSec layers, thereby providing a secured link between the GANC and the FAP.

V. HIGH-LEVEL PROCEDURES

A. Mode Selection in Multi-Mode Terminals

A Generic Access capable UE can support any IP access technology in addition to the UTRAN and possibly GERAN radio interfaces. The UE can be either in the GERAN/UTRAN mode or in GAN mode of operation. The UE can be configured to operate in one of the two modes (i.e., GERAN/UTRAN or GAN) at any given time. There may be a preferred mode of operation that can be configured by the subscriber or by the service provider through various mechanisms, e.g. device management.

On power up, the UE always starts in GERAN/UTRAN mode and executes the normal power-up sequence. The UE in some embodiments executes the power-up sequence as specified in “Non-Access-Stratum functions related to Mobile Station (MS) in idle mode”, 3GPP TS 23.122 standard. Following this, the UE may switch into GAN mode based on mode selection preference determined by user preferences or operator configuration.

The various preferences for the UE that are possible are as follows: GERAN/UTRAN-only, GERAN/UTRAN-preferred, GAN-preferred, and GAN-only. In GERAN/UTRAN-only, the UE RR entity remains in GERAN/UTRAN mode and does not switch to GAN mode. In GERAN/UTRAN-preferred, the UE RR entity is in GERAN/UTRAN mode as long as there is a PLMN available and not forbidden through GERAN/UTRAN. If no allowable PLMN is available through GERAN/UTRAN, and UE has successfully registered with a GAN over the generic IP access network, then the UE switches to GAN mode. When a PLMN becomes available over GERAN/UTRAN and the PLMN is not forbidden, or the UE has de-registered or lost connectivity with the GAN over the generic IP access network, the UE returns to GERAN/UTRAN mode.

In GAN-preferred, when the UE has successfully registered with the GAN over the generic IP access network, the UE switches to GAN mode and stays in this mode as long as the GAN is available. When the UE deregisters, or otherwise loses connectivity with the GAN over the generic IP access network, the UE switches to GERAN/UTRAN mode.

In GAN-only, the UE switches to GAN mode (after initial power up sequence in GERAN/UTRAN mode to obtain cellular network information, but excluding MM and GMM procedures with GERAN/UTRAN core network) and does not switch to GERAN/UTRAN mode. During the initial power up sequence in GERAN/UTRAN mode the UE shall ignore all paging messages received through the GERAN/UTRAN network.

B. PLMN Selection

In some embodiments, there are no changes from the PLMN selection procedures in the NAS layers (MM and above) in the UE, with the exception that in GAN mode the “in VPLMN background scan” is disabled. A GANC can only be connected to one PLMN. The PLMN selection in the NAS layers does not lead to a change of mode between GERAN/UTRAN mode and GAN mode. For a specific instance of PLMN selection, only PLMNs available via GAN or only PLMNs available via GERAN/UTRAN are provided to the NAS layer (i.e., no combination of the PLMNs available via GERAN/UTRAN and GAN).

In the case of a GAN capable UE, some embodiments require a GANC selection process as part of the process of establishing the connectivity between the UE and the GANC. This takes place when, during GAN registration, a GAN capable UE may have a choice among two or more GANC-PLMN pairs indicated by the Default GANC (i.e., in the GA-RC REGISTER REDIRECT message). The GANC selection process takes place while the UE is still in GERAN/UTRAN mode, and before the UE roves into GAN mode. If the current selected PLMN is available via GAN, it shall be selected. If not, the selection of GANC is implementation specific.

If the UE does not have any stored information related to the Serving GANC for the cell or AP to which the UE is currently connected, the UE attempts to register with the Default GANC (always located in the HPLMN) stored in UE. The UE includes an indication, identifying the GANC as the Default GANC in the GA-RC REGISTER REQUEST message.

When a UE attempts to register on the Default GANC including an indication that it is in automatic PLMN selection mode one of the followings happens. If the Default GANC decides to serve the UE, the Default GANC responds with a GA-RC REGISTER ACCEPT message. When the Default GANC decides to redirect the UE to another GANC within the HPLMN, the Default GANC responds with a GA-RC REGISTER REDIRECT message, not including a list of PLMN identities.

When the Default GANC decides to redirect the UE to a PLMN that is not the HPLMN, the Default GANC responds with a GA-RC REGISTER REDIRECT message and includes a list of PLMNs that may provide GAN service to the UE in its current location. The list contains one or more PLMN identities along with the identities of their associated GANC and SEGW nodes (either in IP address or FQDN format). Following the GANC selection process, the GA-RC entity in the UE attempts to register on the associated GANC.

If at any time the user wishes to perform manual PLMN selection or a “User reselection” irrespective of whether the UE is in manual or automatic PLMN selection mode, the UE sends a GA-RC REGISTER REQUEST message to the Default GANC, including an indication that it is in manual PLMN selection mode. The Default GANC is not allowed to accept the registration and responds with a GA-RC REGISTER REDIRECT message and includes a list of PLMNs that may provide GAN service to the UE in its current location.

When the UE includes the identity of the current serving GSM network in the GA-RC REGISTER REQUEST message, the Default GANC uses this to identify the list of PLMNs to send to the UE in the response message.

After successful registration with a serving GANC, the UE does not store the PLMN list. The UE does not use the PLMN list, provided to the UE during the registration procedure, for background scanning. A UE cannot use GA in a VPLMN unless the HPLMN supports and authorizes GA.

C. Re-Selection Between GERAN/UTRAN and GAN Modes

1. Rove-in (from GERAN/UTRAN Mode to GAN Mode)

This procedure is applicable only when GAN service is available, a UE is not in NC2 mode (applicable if the UE is in GERAN mode and as defined in “Radio subsystem link control”, 3GPP TS 45.008 standard) and has a UE preference for GAN-only, GAN-preferred or, if no allowable PLMN is available through GERAN/UTRAN, for GERAN/UTRAN-preferred.

Following successful GAN registration, the access mode in the UE is switched to GAN mode. The GA-CSR entity in the UE provides the NAS-related system information received in the GAN Registration Procedure to the NAS layers. The NAS considers the GANC-allocated cell identity as the current serving cell.

While in GAN mode, GERAN-RR and UTRAN RRC entities are detached from the RR-SAP in the UE. As a result the entities do not: (1) inform NAS about any GERAN/UTRAN cell re-selection and/or the change of system information of the current camping cell, (2) inform NAS about any newly found PLMN over GERAN or UTRAN, and (3) act on any paging request message received over GERAN or UTRAN.

2. Rove-Out (from GAN Mode to GERAN/UTRAN Mode)

This procedure is applicable when the UE detaches from the generic IP access network, and its mode selection is GAN-preferred or GERAN/UTRAN-preferred. When the UE detaches from the generic IP access network, depending on prevailing circumstances the UE may be able to deregister first with the GANC.

For the GAN-preferred and GERAN/UTRAN-preferred mode selections, the UE detaches the GA-CSR entity from the RR-SAP and re-attaches the GERAN-RR or UTRAN RRC entity to the RR-SAP and restores normal GERAN-RR or UTRAN RRC functionality. For the GAN-only mode selection, GA-CSR remains attached to the NAS and the UE stays in GAN mode (i.e., in “No Service” condition).

D. GAN Registration Related Procedures

1. Discovery and Registration for Generic Access

The Discovery and Registration procedures are applicable only if the UE preference is operating in GAN-only, GAN-preferred or, if no allowable PLMN is available through GERAN/UTRAN, in GERAN/UTRAN-preferred mode.

Once the UE has established a connection to the generic IP access network, the UE determines the appropriate GANC-SEGW to connect to, by completing the Discovery Procedure to the Provisioning GANC in the HPLMN of the UE. The Provisioning GANC provides the address of the Default GANC in the HPLMN of the UE, to which the UE can register.

The UE attempts to register on the Default GANC provided by the Provisioning GANC during the Discovery procedure, by completing the Registration Procedure. The Default GANC may accept the Registration; redirect the UE to another GANC; or reject the Registration.

a) Security Gateway Identification

The USIM of the UE contains the FQDN (or IP address) of the Provisioning GANC and the associated SEGW or the UE derives this information based on information in the USIM. When the UE does not have any information about other GANCs and associated SEGW stored, then the UE completes the Discovery procedure towards the Provisioning GANC. As part of the Registration Procedure, the Default GANC can indicate whether this GANC and SEGW address or the address of a GANC that the UE is being redirected to, may be stored by the UE.

The UE can also store Serving GANC information for Serving GANCs with which the UE was able to complete a successful registration procedure. The default GANC is in control of whether the UE is allowed to store Serving GANC information. When there is no GERAN/UTRAN coverage in the AP location, the stored Serving GANC information is associated with the AP-ID. When there is GERAN/UTRAN coverage in the AP location, the stored Serving GANC information is associated with the GSM CGI or LAI or UTRAN CI. The stored Serving GANC information is: (1) serving SEGW FQDN or IP address following successful registration, (2) serving GANC FQDN or IP address following successful registration, and (3) optionally, Serving GANC TCP port following successful registration and if returned from the network. Different embodiments store different number of such entries in the UE is implementation specific. Only the last successfully registered GANC association is stored when the Default GANC indicates that the UE is allowed to store these addresses. A UE may preferentially join a generic IP access network point of attachment whose association with a Serving GANC has been stored in memory.

On connecting to the generic IP access network, when the UE has a stored Serving GANC for the AP-ID or the GERAN/UTRAN cell, the UE attempts to register with the associated Serving GANC in its memory. The GANC may still reject the UE for any reason even though it may have served the UE before. The UE deletes from its stored list the address of the Serving GANC on receiving a registration reject or if the registration fails for any other reason (e.g., not receiving any response).

If the UE does not receive a response to the Registration Request sent to the Serving GANC (and which is not the Default GANC), the UE re-attempt to register with the Default GANC. If the UE does not receive a response to the registration request sent to the Default GANC, it attempts the discovery procedure with the Provisioning GANC to obtain a new Default GANC.

In the case when a UE is attempting to register or discover a GANC after failing to register on a GANC, the UE provides in the Registration or Discovery procedure an indication that the UE has attempted to register on another GANC, the failure reason, and the GANC and SEGW addresses of the failed registration. When the UE connects to a generic IP access network, for which the UE does not have a stored Serving GANC in it's memory, the UE attempt to register with the Default GANC.

b) GANC Capabilities

GANC specific information is transferred to the UE on successful registration.

c) UE Capabilities

GAN specific capabilities of the UE are transferred to the GANC during registration.

d) Required GAN Services

The UE may request which GAN services it requires from the GANC as part of the Registration procedures.

e) GAN Mode Selection

The UE (i.e., with Iu-mode GAN support) transfers its GAN Mode Support information to the GANC during Discovery and Registration procedures; i.e., in the GAN Classmark IE. GAN Mode Support options are A/Gb mode supported, Iu mode supported, or both modes supported. When no GAN Mode Support information is received, the GANC assumes that the UE supports A/Gb mode operation only.

The provisioning GANC may use the received GAN Mode Support information to assign the UE to an appropriate default GANC (e.g., if separate A/Gb mode and Iu-mode GANCs are deployed in the network) or to an appropriate TCP port on the default GANC (e.g., if separate TCP ports are used for A/Gb mode and Iu-mode GAN service). The Iu-mode capable GANC also indicates the GAN mode to use for the current session in the GAN Mode Indicator IE; this allows the UE to determine the Iu-mode capability of the Home PLMN.

Table 1 enumerates the discovery handling for the various combinations of UE and Home PLMN GAN mode capabilities.

TABLE 1


GAN Mode Selection procedures associated with GAN Discovery

UE GAN Mode

Home PLMN GAN Mode Capabilities

Capabilities	A/Gb only	Iu only	Both

A/Gb only	GANC: Handle as	GANC: No GAN	GANC: No GAN
	normal A/Gb mode	Mode Support	Mode Support
	discovery	information provided	information provided
	UE: Proceed with	or A/Gb mode (only)	or A/Gb mode (only)
	A/Gb mode	indicated by UE,	indicated by UE,
	registration	therefore Reject	therefore handle as
		(Unspecified)	normal A/Gb mode
		UE: Retry on next	discovery. Assign UE
		power-on	to A/Gb-capable
			GANC.
			UE: Proceed with
			A/Gb mode
			registration
Iu only	GANC: Handle as	GANC: Iu Mode	GANC: Iu Mode
	normal A/Gb mode	Support (only)	Support (only)
	discovery	indicated by UE,	indicated by UE,
	UE: No GAN Mode	therefore accept and	therefore accept and
	Selection provided by	send GAN Mode	send GAN Mode
	GANC, therefore abort	Indicator = Iu	Indicator = Iu. Assign
	GAN operation and	UE: Proceed with Iu	UE to Iu-capable
	retry on next power-on	mode registration	GANC.
			UE: Proceed with Iu
			mode registration
Both	GANC: Handle as	GANC: Support for	GANC: Support for
	normal A/Gb	both modes indicated	both modes indicated
	discovery	by UE, therefore	by UE, therefore
	UE: No GAN Mode	accept and send GAN	accept and send GAN
	Selection provided by	Mode Indicator = Iu	Mode Indicator = Iu.
	GANC, therefore	UE: Proceed with Iu	Assign UE to Iu-
	proceed with Iu mode	mode registration	capable GANC.
	registration (Note 1)		UE: Proceed with Iu
			mode registration

Note:
As described in Table 2 below, the result of Iu mode registration of a A/Gb-capable UE on a A/Gb-capable GANC is that the UE is placed in A/Gb mode.

In some embodiments, the default or serving GANC uses the received GAN Mode Support information to redirect the UE to a different GANC or a different TCP port on the current GANC. The Iu-mode capable GANC also indicates the GAN mode to use for the current session in the GAN Mode Indicator IE.

Table 2 enumerates the registration handling for the various combinations of UE and Home PLMN GAN mode capabilities.

TABLE 2


GAN Mode Selection procedures associated with GAN Registration

UE GAN Mode

Default/Serving GANC GAN Mode Capabilities

Capabilities	A/Gb only	Iu only	Both

A/Gb only	GANC: Handle as	GANC: No GAN	GANC: No GAN
	normal A/Gb mode	Mode Support	Mode Support
	registration	information provided	information provided
	UE: Proceed per A/Gb	or A/Gb mode (only)	or A/Gb mode (only)
	mode GAN procedures	indicated by UE,	indicated by UE,
		therefore Reject	therefore handle as
		(Invalid GANC)	normal A/Gb mode
		UE: Attempt	registration. If
		registration with	required, redirect UE
		Default GANC or re-	to A/Gb-capable
		discovery (per A/Gb	GANC.
		mode GAN	UE: Proceed per A/Gb
		procedures)	mode GAN procedures
Iu only	GANC: Handle as	GANC: Iu Mode	GANC: Iu Mode
	normal A/Gb mode	Support (only)	Support (only)
	registration	indicated by UE,	indicated by UE,
	UE: No GAN Mode	therefore accept and	therefore accept and
	Selection provided by	send GAN Mode	send GAN Mode
	GANC, therefore	Indicator = Iu	Indicator = Iu.
	Deregister and treat as	UE: Proceed per Iu	UE: Proceed per Iu
	register reject (Invalid	mode GAN procedures	mode GAN procedures
	GANC)
Both	GANC: Handle as	GANC: Support for	GANC: Support for
	normal A/Gb	both modes indicated	both modes indicated
	registration	by UE, therefore	by UE, therefore
	UE: No GAN Mode	accept and send GAN	accept and send GAN
	Selection provided by	Mode Indicator = Iu	Mode Indicator = Iu or
	GANC, therefore	UE: Proceed per Iu	A/Gb (seeNote 1
	proceed per A/Gb	mode GAN procedures	below). If required,
	mode GAN procedures		redirect UE to Iu or
			A/Gb-capable GANC.
			UE: Proceed per Iu or
			A/Gb mode GAN
			procedures

Note 1:
The GANC's choice of Iu-mode versus A/Gb-mode may be based on other information received in the GAN registration message from the UE, information stored in the GANC, and on operator (i.e., service provider) policy; e.g., if the GSM RR/UTRAN RRC State IE indicates that the UE is in GERAN Dedicated mode, the UE location is an area without UTRAN coverage and the operator wants to minimize inter-RAT handovers, the GANC may direct the UE to use A/Gb mode.

f) Discovery Procedure

When a UE supporting GAN first attempts to connect to a GAN, the UE needs to identify the Default GANC. Each GAN capable UE can be configured with the FQDN (or IP address) of the Provisioning GANC and the associated SEGW or the UE can derive this FQDN based on information in the USIM (see “Numbering, addressing and identification”, 3GPP TS 23.003 standard). The UE first connects to a Provisioning GANC-SEGW and GANC in the HPLMN of the UE, by establishing a secure IPSec tunnel and a TCP connection using the provisioned or derived addresses. The UE obtains the FQDN or IP address of the Default GANC in the HPLMN and the associated SEGW, through the Discovery procedure.

If no GERAN/UTRAN coverage is available when a UE connects to the GANC for GAN service, then the GANC cannot necessarily determine the location of the UE for the purposes of assigning the UE to the correct serving GANC (e.g., to enable handover and location-based services). The GANC permits the operator to determine the service policy in this case; e.g., the operator could provide service to the user with certain limitations (possibly with a user interface indication on the UE). When the UE initiates the Discovery/Registration procedures and no GERAN/UTRAN coverage is available, the GANC may have insufficient information to correctly route subsequent emergency calls.

FIG. 23 illustrates the Discovery procedure in some embodiments. The figure shows different messages exchanges between theUE2305,DNS2310, theprovisioning GANC2315, thesecurity gateway SEGW2320 associated with theprovisioning GANC2315, and theDNS server2325 associated with theprovisioning GANC2315. In the description below it is assumed that theUE2305 has a mode selection of GAN-only or GAN-preferred or GERAN/UTRAN-preferred and that the UE has already connected to the generic IP access network. Different embodiments deem different signal levels as sufficient for triggering the GAN Discovery and Registration procedures. The following steps are taken during Discovery procedure in some embodiments.

As shown inFIG. 23, when theUE2305 has a provisioned or derived FQDN of the Provisioning SEGW, the UE performs (in Step1) a DNS query (via the generic IP access network interface) to resolve the FQDN to an IP address. When the UE has a provisioned IP address for the Provisioning SEGW, the DNS step is omitted. Next, theDNS Server2310 returns (in Step2) a response including the IP Address of theProvisioning SEGW2320.

As shown, theUE2305 establishes (in Step3) a secure tunnel to theProvisioning SEGW2320. When theUE2305 has a provisioned or derived FQDN of theProvisioning GANC2315, theUE2305 performs (in Step4) a DNS query (via the secure tunnel) to theDNS server2325 associated with theprovisioning GANC2315 to resolve the FQDN to an IP address. When theUE2305 has a provisioned IP address for the Provisioning GANC, the DNS step will be omitted. TheDNS Server2325 returns (in Step5) a response including the IP Address of theProvisioning GANC2315.

TheUE2305 sets up a TCP connection to a well-defined port on theProvisioning GANC2315. It then queries (in Step6) theProvisioning GANC2315 for the Default GANC, using GA-RC DISCOVERY REQUEST. The message contains: (1) Cell Info: Either current camping UTRAN/GERAN cell ID or the last LAI where the UE successfully registered, along with an indicator stating which one it is, (2) Generic IP access network attachment point information: AP-ID, as defined in Identifiers in GAN, sub-section VII, below, (3) UE Identity: IMSI, and (4) GAN Classmark: Including indications of A/Gb Mode supported and Iu Mode supported.

Next, theProvisioning GANC2315 returns (in Step7) the GA-RC DISCOVERY ACCEPT message, using the information provided by the UE (e.g. the cell ID), to provide the FQDN or IP address of the Default GANC and its associated Default SEGW. This is done so the UE is directed to a “local” Default GANC in the HPLMN to optimize network performance. The GANC Port that the UE must use for registration may be included. The GAN Mode Indicator may be included as described in GAN Mode Section, sub-section above.

When theProvisioning GANC2315 cannot accept the GA-RC DISCOVERY REQUEST message, it returns (in Step8) a GA-RC DISCOVERY REJECT message indicating the reject cause. The secure IPSec tunnel to theProvisioning SEGW2320 is released (in Step9). It is possible to reuse the same IPSec tunnel for GAN Registration procedures. In this case the IPSec tunnel is not released.

g) Registration Procedure—Normal Case

Following the Discovery procedure the UE establishes a secure tunnel with the security gateway of the Default GANC, provided by the Provisioning GANC in the Discovery procedure, and attempts to register with the Default GANC. The Default GANC may become the Serving GANC for that connection by accepting the registration, or the Default GANC may redirect a UE performing registration to a different Serving GANC.

GANC redirection may be based on information provided by the UE during the Registration procedure, operator chosen policy or network load balancing. The GAN Registration procedure serves the following functions: (1) Ensures the UE is registered to the appropriate GANC entity; i.e., with use of the redirection process, (2) Informs the GANC that the UE is now connected through a generic IP access network and is available at a particular IP address. The GANC maintains the registration context for the purposes of (for example) mobile-terminated calling, (3) Provides the UE with the operating parameters associated with the GAN service. The “System Information” message content that is applicable to the GAN cell is delivered to the UE during the GAN registration process. This enables the UE to switch to GAN mode, and following the Registration procedure trigger NAS procedures with the core network (such as Location/Routing Area Update, mobile originated calls, mobile terminated calls, etc.), and (4) Enables the UE to request which GAN services are required.

FIG. 24 illustrates Registration procedure in some embodiments. The figure shows different messages exchanges between theUE2405,DNS2410, theprovisioning GANC2415, thesecurity gateway SEGW2420 associated with theprovisioning GANC2415, and theDNS server2425 associated with theprovisioning GANC2415. The following steps are done during Registration procedure.

As shown inFIG. 24, when theUE2405 was provided the FQDN of the Default or Serving SEGW, the UE performs (in Step1) a DNS query (via the generic IP access network interface) to resolve the FQDN to an IP address. When the UE has a provisioned IP address for the SEGW, the DNS step is omitted. TheDNS Server2410 returns (in Step2) a response.

As shown, theUE2405 sets up (in Step3) a secure IPSec tunnel to theSEGW2420. This step may be omitted if an IPSec tunnel is being reused from an earlier Discovery or Registration. When theUE2405 was provided the FQDN of the Default or Serving GANC, the UE then performs (in Step4) a DNS query (via the secure tunnel) to resolve the FQDN to an IP address. When the UE has an IP address for the GANC, the DNS step is omitted. Next, theDNS Server2425 returns (in Step5) a response.

TheUE2405 then sets up a TCP connection to a TCP port on the GANC. The TCP port can either be a well-known port or one that has been earlier received from the network during Discovery or Registration. TheUE2405 attempts (in Step6) to register on the GANC by transmitting the GA-RC REGISTER REQUEST. The message includes: (1) Cell Info: Either current camping UTRAN/GERAN cell ID, or last LAI where the UE successfully registered, along with an indicator stating which one it is, (2) Generic IP access network attachment point information: AP-ID, as defined in Identifier in GAN, Section VII, below, (3) UE Identity: IMSI, (4) UE Capability Information, (5) GAN Services Required, (6) GAN Classmark: Including indications of A/Gb Mode supported, Iu Mode supported.

When theGANC2415 accepts the registration attempt, theGANC2415 responds (in Step7) with a GA-RC REGISTER ACCEPT. In this case the TCP connection and the secure IPSec tunnel are not released and are maintained as long as the UE is registered to this GANC.

The GA-RC REGISTER ACCEPT message includes (1) GAN Capability Information and (2) GAN specific system information which includes (a) GAN Mode Indicator:

A/Gb Mode GAN or Iu Mode GAN, (b) Cell description of the GAN cell, (c) Location-area identification comprising the mobile country code, mobile network code, and location area code corresponding to the GAN cell, (d) Cell identity identifying the cell within the location area corresponding to the GAN cell, and (e) Applicable system timer values (e.g., for the application-level keep alive message transmission interval, see Keep Alive sub-section, below)

Alternatively, theGANC2415 may reject the request. In this case, theGANC2415 responds (in Step8) with a GA-RC REGISTER REJECT indicating the reject cause. The TCP connection and the secure IPSec tunnel are then released.

Alternatively, if theGANC2415 decides to redirect the UE to (another) Serving GANC, theGANC2415 responds (in Step9) with a GA-RC REGISTER REDIRECT providing the FQDN or IP address of the target Serving GANC and the associated SEGW, and the GAN Mode Indicator if the GANC requires that a particular mode be used with the Serving GANC (e.g., if the GANC knows that the Serving GANC supports only A/Gb mode GAN). In this case the TCP connection is released and the secure IPSec tunnel is optionally released (in Step10) depending on if the network indicates that the same IPSec tunnel can be reused for the next registration. The GA-RC REGISTER REDIRECT message may contain: (1) a single Serving SEGW and GANC address or (2) a list of PLMN identities and associated Serving SEGW and GANC addresses. The message also may contain an Indication of whether GANC address(es) can be stored in the UE for future use.

a) Registration Procedure—Abnormal Cases

When the Serving GANC rejects the Register request and does not provide redirection to another Serving GANC, the UE re-attempts Registration to the Default GANC including a cause that indicates the failed registration attempt and the Serving GANC and SEGW with which the Register request failed. The UE also deletes all stored information about this Serving GANC.

When the Default GANC rejects a Registration Request and is unable to provide redirection to suitable Serving GANC, the UE may re-attempt the Discovery procedure to the Provisioning GANC (including a cause indicating the failed registration attempt and the Default GANC provided in the last Discovery procedure). The UE also deletes all stored information about the Default GANC.

2. De-Registration

FIG. 25 illustrates De-Registration initiated by theUE2505 in some embodiments. The GA-RC De-Registration procedure allows theUE2505 to explicitly inform theGANC2510 that it is leaving GAN mode (e.g., when it detaches from the generic IP access network), by sending (in Step1) a GA-RC DEREGISTER message to theGANC2510, allowing theGANC2510 to free resources that it assigned to theUE2505. TheGANC2510 also supports “implicit GAN de-registration”, when the TCP connection to the UE is abruptly lost.

FIG. 26 illustrates De-Registration initiated by theGANC2610 in some embodiments. As shown, theGANC2610 can autonomously release the UE registration context, and send (in Step1) a GA-RC DEREGISTER message to theUE2605. Alternatively, theGANC2610 can implicitly deregister theUE2605 by closing the TCP connection with the UE. At power-down the GA-RC sublayer of the UE ensures that the UE explicitly detaches from the network, where possible, before completing the GA-RC De-Registration procedure.

3. Registration Update

FIG. 27 illustrates Registration Update in some embodiments. The GA-RC Registration Update procedure allows theUE2705 to update information in theGANC2710 regarding changes to the identity of the overlapping GERAN cell or changes to the generic IP access network point of attachment. As shown, theUE2705 sends (in Step1) a GA-RC REGISTER UPDATE UPLINK message to theGANC2710 carrying the updated information. This may result in theUE2705 being redirected to another serving GANC, or being denied service; e.g., due to operator policy.

When theUE2705 detects UTRAN/GERAN coverage after reporting no coverage during GAN registration, the UE sends the GA-RC REGISTER UPDATE UPLINK to the GANC with the updated information. Whenever the generic IP access network point of attachment changes, the UE sends a GA-RC REGISTER UPDATE UPLINK to the GANC with the updated generic IP access network point of attachment information. When the UE requires to update the GANC with a new list of GAN Services required, then the UE sends GA-RC REGISTER UPDATE UPLINK message to the GANC including the new GAN Services Required list.

TheGANC2710 may optionally send (in Step2) the GA-RC REGISTER REDIRECT when it decides to redirect the UE based on updated information. TheGANC2710 may also optionally deregister theUE2705 on receiving an update by sending (in Step3) GA-RC DEREGISTER to the UE.

FIG. 28 illustrates Registration Update Downlink procedure in some embodiments. The GAN Registration Update procedure also allows theGANC2810 to update the GAN system information in theUE2805, if needed, by sending (in Step1) a GA-RC REGISTER UPDATE DOWNLINK message to the UE carrying the updated information.

4. Keep Alive

FIG. 29 illustrates the Keep Alive process in some embodiments. The Keep Alive process is a mechanism between the peer GA-RC entities to indicate that the UE is still registered to the GANC. Using periodic transmissions (in Step1) of the GA-RC KEEP ALIVE message theUE2805 in turn determines that theGANC2810 is still available using the currently established lower layer connection.

5. Cell Broadcast Information

FIG. 30 illustrates the Cell Broadcast Information mechanism of some embodiments. The Cell Broadcast Information is a mechanism between the peer GA-RC entities, allowing the GANC to pass the UE information relating to the Cell Broadcast Services. TheUE3005 includes GAN Service Required information in the GA-RC REGISTER REQUEST and GA-RC REGISTER UPDATE UPLINK messages passed to the GANC, indicating that the UE requires the Cell Broadcast Service. TheGANC3010 then passes (in Step1) the required information to theUE1105 in the GA-RC CELL BROADCAST INFO message.

E. Authentication

The Up interface supports the ability to authenticate the UE with the GANC (for the purposes of establishing the secure tunnel) using GSM or UMTS credentials. Authentication between UE and GANC is performed using EAP-SIM or EAP-AKA within IKEv2.

F. Encryption and Integrity Protection

All control and user plane traffic over the Up interface is sent through the pair of IPSec ESP tunnel mode security associations (one for each direction) that are established during the establishment of the IKEv2 security association. Encryption and integrity protection are via the negotiated cryptographic algorithms, based on core network policy, enforced by the GANC-SEGW.

G. GA-CSR Connection Handling

The Iu-mode GAN GA-CSR connection is a logical connection between the UE and the GANC for the CS domain. A GA-CSR connection is established when the upper layers in the UE request the establishment of a CS domain signaling connection and the UE is in GA-CSR-IDLE state; i.e., no GA-CSR connection exists. When a successful response is received from the network, GA-CSR replies to the upper layer that the CS domain signaling connection is established and the UE has entered the equivalent of the RRC connected mode (i.e., the GA-CSR-CONNECTED state).

1. GA-CSR Connection Establishment

FIG. 31 illustrates successful and unsuccessful establishment of the GA-CSR Connection in some embodiments. As shown, theUE3105 initiates GA-CSR connection establishment by sending (in Step1) the GA-CSR REQUEST message to theGANC3110. This message contains the Establishment Cause indicating the reason for GA-CSR connection establishment.

When GANC determines that the connection request can be accepted, theGANC3110 signals the acceptance of the connection request to theUE3105 by sending (in Step2) the GA-CSR REQUEST ACCEPT and the UE enters the GA-CSR-CONNECTED state. On the other hand, when the GANC determines that the GA-CSR connection request has to be rejected, theGANC3110 sends (in Step3) a GA-CSR REQUEST REJECT to theUE3105 indicating the reject cause, completing the procedure.

2. GA-CSR Connection Release

FIG. 32 illustrates release of the logical GA-CSR connection between the UE and the GANC in some embodiments. As shown, theMSC3215 indicates to theGANC3210 to release the CS resources allocated to the UE, by sending (in Step1) the RANAP Iu Release Command message to theGANC3210.

Next, theGANC3210 confirms (in Step2) resource release toMSC3215 using the Iu Release Complete message. TheGANC3210 then commands (in Step3) theUE3205 to release resources, using the GA-CSR RELEASE message. Finally, theUE3205 confirms (in Step4) resource release to the GANC using the GA-CSR RELEASE COMPLETE message and the GA-CSR state in the UE changes to GA-CSR-IDLE.

H. CS Security Mode Control

FIG. 33 illustrates the message flow for security mode control in some embodiments. As shown, theMSC3315 sends (in Step1) the RANAP Security Mode Command message toGANC3310. This message contains the integrity key (IK) and allowed algorithms, and optionally the encryption key (CK) and allowed algorithms.

Next, theGANC3310 sends (in Step2) the GA-CSR SECURITY MODE COMMAND message to theUE3305. This message indicates the integrity protection and encryption settings (i.e., that are applicable after relocation to UTRAN), and a random number. The UE stores the information for possible future use after a relocation to UTRAN.

Next, theUE3305 computes a MAC based on the random number, the UE IMSI and the integrity key calculated by the UE. The MAC or “message authentication code” allows the GANC to verify that the UE has been able to calculate the same integrity key value as the GANC received from the MSC, thereby preventing certain “man-in-the-middle” security attacks. TheUE3305 then sends (in Step3) the GA-CSR SECURITY MODE COMPLETE message to theGANC3310 to signal its selected algorithm and the computed MAC.

TheGANC3310 then verifies the MAC using the random number, the UE IMSI and the integrity key provided by the MSC inStep1. When the GANC verifies the MAC to be correct (i.e., the GANC-calculated MAC is the same as the UE-calculated MAC) it sends (in Step4) the Security Mode Complete message to theMSC3315. The MAC proves that the identity that is authenticated to the GANC is the same as the identity authenticated to the core network.

I. CS NAS Signaling Procedures

After GA-CSR connection establishment, NAS signaling may be transferred from MSC-to-UE and from UE-to-MSC.

1. MSC-to-UE NAS Signaling

FIG. 34 illustrates MSC-to-UE NAS signaling in some embodiments. As shown, for MSC-to-UE NAS signaling, theMSC3415 sends (in Step1) a NAS PDU to the GANC via the RANAP Direct Transfer message. TheGANC3410 encapsulates the NAS PDU within a GA-CSR DL DIRECT TRANSFER message and forwards (in Step2) the message to theUE3405 via the existing TCP connection.

2. UE-to-MSC NAS Signaling

FIG. 35 illustrates UE-to-MSC NAS signaling in some embodiments. As shown, theUE3505 receives a request from the NAS layer to transfer an uplink NAS PDU. Assuming that the required signaling connection already exists, theUE3505 encapsulates the NAS PDU within a GA-CSR UL DIRECT TRANSFER message and sends (in Step1) the message to theGANC3510. TheGANC3510 relays (in Step2) the received message to theMSC3515 via the RANAP Direct Transfer message.

J. Mobile Originated CS Call

1. GANC Terminates Iu UP Protocol

FIG. 36 illustrates steps performed during a mobile originated CS call in some embodiments. The procedure assumes that the UE is in GAN mode; i.e., it has successfully registered with the GANC and GA-CSR is the serving RR entity for CS services in the UE. It also assumes that no GA-CSR signaling connection exists between the UE and GANC (i.e., GA-CSR-IDLE state). As shown, the GA-CSR Connection Establishment procedure is performed (in Step1). In some embodiments, this procedure is performed as described in GA-CSR Connection Establishment sub-section, above. Next, theUE3605 sends the CM Service Request message to theGANC3610 within the GA-CSR UL DIRECT TRANSFER message.

Next, theGANC3610 establishes an SCCP connection to theMSC3615 and forwards (in Step3) the NAS PDU (i.e., the CM Service Request message) to theMSC3615 using the RANAP Initial UE Message. The message includes the Domain Indicator set to value ‘CS domain’. Subsequent NAS messages between the UE and MSC will be sent between GANC and MSC using the RANAP Direct Transfer message.

TheMSC3615 may optionally authenticate (in Step4) the UE using standard UTRAN authentication procedures. TheMSC3615 may optionally initiate (in Step5) the Security Mode Control procedure described in CS Security Mode Control sub-section, above. TheUE3605 sends (in Step6) the Setup message providing details on the call to the MSC and its bearer capability and supported codecs. This message is contained within the GA-CSR UL DIRECT TRANSFER between the UE and the GANC. The GANC forwards the Setup message to the MSC.

Next, theMSC3615 indicates (in Step7) it has received the call setup and it will accept no additional call-establishment information using the Call Proceeding message to the GANC. The GANC forwards (in Step7) this message to the UE in the GA-CSR DL DIRECT TRANSFER message.

TheMSC3615 requests (in Step8) theGANC3610 to assign call resources using the RANAP RAB Assignment Request message. TheMSC3615 includes the RAB-ID, the CN Transport Layer Address and the CN Iu Transport Association for user data, and an indication that Iu UP support mode is required, among other parameters.

TheGANC3610 then sends (in Step9) the GA-CSR ACTIVATE CHANNEL message to theUE3605 including bearer path setup information such as: (1) Channel mode, (2) Multi-rate codec configuration, (3) UDP port & the IP address for the uplink RTP stream, and (4) Voice sample size.

Next, theUE3605 sends (in Step10) the GA-CSR ACTIVATE CHANNEL ACK to theGANC3610 indicating the UDP port for the downlink RTP stream. Since Iu UP support mode is indicated by the MSC instep8, theGANC3610 sends (in Step11) the Iu UP INITIALIZATION packet to the MSC.

In response, the MSC responds (in Step12) with the Iu UP INITIALISATION ACK packet. TheGANC3610 signals (in Step13) the completion of the RAB establishment to theUE3605 with the GA-CSR ACTIVATE CHANNEL COMPLETE message. Alternatively, Steps11 and12 may occur beforeStep9.

TheGANC3610 signals to theMSC3615 that the RAB has been established by sending (in Step14) a RANAP RAB Assignment Response message. TheMSC3615 signals to theUE3505, with the Alerting message, that the called party is ringing. The message is transferred (in Step15) to theGANC3610 and GANC forwards (in Step15) the message to theUE3605 in the GA-CSR DL DIRECT TRANSFER. When the UE has not connected the audio path to the user, it generates ring back to the calling party. Otherwise, the network-generated ring back will be returned to the calling party.

Next, theMSC3615 signals that the called party has answered, via the Connect message. The message is transferred (in Step16) to theGANC3610 and GANC forwards (in Step16) the message to the UE in the GA-CSR DL DIRECT TRANSFER. The UE connects the user to the audio path. If the UE is generating ring back, it stops and connects the user to the audio path.

TheUE3605 then sends (in Step17) the Connect Ack message in response, and the two parties are connected for the voice call. This message is contained within the GA-CSR UL DIRECT TRANSFER between the UE and the GANC. The GANC forwards the Connect Ack message to the MSC. At this time, bi-directional voice traffic flows (in Step18) between theUE3605 andMSC3615 through theGANC3610.

2. UE Terminates Iu UP Protocol

Some embodiments utilize an alternative procedure for a mobile originated CS call.FIG. 37 illustrates steps performed during a mobile originated CS call in these embodiments. The procedure assumes that the UE is in GAN mode; i.e., it has successfully registered with the GANC and GA-CSR is the serving RR entity for CS services in the UE. It also assumes that no GA-CSR signaling connection exists between the UE and GANC (i.e., GA-CSR-IDLE state).Steps1 to8 are performed the same as described forsteps1 to8 shown inFIG. 36 above and are not repeated for simplicity.

Since Iu UP support mode is indicated by the MSC in step8 (as described in reference withFIG. 36), the GANC indicates (in Step9) that Iu UP support mode is required in the GA-CSR ACTIVATE CHANNEL message, and theUE3705 sends (in Step10) the Iu UP INITIALIZATION packet to theMSC3715. In response, theMSC3715 responds (in Step11) with the Iu UP INITIALISATION ACK packet. Next, theUE3705 sends (in Step12) the GA-CSR ACTIVATE CHANNEL ACK to theGANC3710.

TheGANC3710 signals to theMSC3715 that the RAB has been established by sending (in Step13) a RANAP RAB Assignment Response message. TheGANC3710 also sends (in Step14) a GA-CSR ACTIVATE CHANNEL COMPLETE message to theUE3705.Steps15 to18 are performed the same as described forsteps15 to18 shown inFIG. 36 above and are not repeated for simplicity.

K. Mobile Terminated CS Call

FIG. 38 illustrates steps performed during a mobile terminated CS call in some embodiments. The description of the procedure assumes that the UE is in GAN mode; i.e., it has successfully registered with the GANC and GA-CSR is the serving RR entity for CS services in the UE. It also assumes that no GA-CSR signaling connection exists between the UE and GANC (i.e., the UE is in the GA-CSR-IDLE state). When a mobile-terminated call arrives at theMSC3815, as shown inFIG. 38, theMSC3815 sends (in Step1) a RANAP Paging message to theGANC3810 identified through the last Location Update received by it and includes the TMSI if available. The IMSI of the mobile being paged is always included in the request.

Next, theGANC3810 identifies the UE registration context using the IMSI provided by theMSC3815. It then pages (in Step2) theUE3805 using the GA-CSR PAGING REQUEST message. The message includes the TMSI, if available in the request from the MSC; else it includes only the IMSI of the UE.

TheUE3805 responds with a GA-CSR PAGING RESPONSE. The UE transitions to the GA-CSR CONNECTED state. TheGANC3810 establishes an SCCP connection to theMSC3815. TheGANC3810 then forwards (in Step4) the paging response to theMSC3815 using the RANAP Initial UE Message. Subsequent NAS messages between the UE and core network will be sent between GANC and MSC using the RANAP Direct Transfer message.

TheMSC3815 may optionally authenticate (in Step5) theUE3805 using standard UTRAN authentication procedures. The MSC may optionally update (in Step6) the security configuration in the UE, via the GANC, as described in CS Security Mode Control sub-section above.

TheMSC3815 then initiates (in Step7) call setup using the Setup message sent to the UE via GANC. GANC forwards (in Step7) this message to theUE3805 in the GA-CSR DL DIRECT TRANSFER message.

Next, theUE3805 responds with Call Confirmed using the GA-CSR UL DIRECT TRANSFER after checking it's compatibility with the bearer service requested in the Setup and modifying the bearer service as needed. When the Setup included the signal information element, the UE alerts the user using the indicated signal, otherwise the UE alerts the user after the successful configuration of the user plane. TheGANC3810 forwards (in Step8) the Call Confirmed message to theMSC3815.

Next, theMSC3815 initiates the assignment procedure with theGANC3810, which triggers (in Step9) the setup of the RTP stream (voice bearer channel) between the GANC and UE, same as steps8-14 in the mobile originated CS call scenario described above.

TheUE3805 then signals (in Step10) that it is alerting the user, via the Alerting message contained in the GA-CSR UL DIRECT TRANSFER. The GANC forwards (in Step10) the Alerting message to the MSC. The MSC sends a corresponding alerting message to the calling party.

TheUE3805 then signals (in Step11) that the called party has answered, via the Connect message contained in the GA-CSR UL DIRECT TRANSFER. TheGANC3810 forwards (in Step11) the Connect message to theMSC3815. The MSC sends a corresponding Connect message to the calling party and through connects the audio. The UE connects the user to the audio path.

Next, theMSC3815 acknowledges (in Step12) via the Connect Ack message to theGANC3810. GANC forwards (in Step12) this message to theUE3805 in the GA-CSR DL DIRECT TRANSFER. The two parties on the call are connected on the audio path. At this time, bi-directional voice traffic flows (in Step13) between the UE and MSC through the GANC.

L. CS Call Clearing

FIG. 39 illustrates call clearing initiated by the UE in some embodiments. As shown, theUE3905 sends (in Step1) the Disconnect message to theMSC3915 to release the call. This message is contained in the GA-CSR UL DIRECT TRANSFER message betweenUE3905 andGANC3910. TheGANC3910 forwards (in Step1) the Disconnect message to the MSC (i.e., using the RANAP Direct Transfer message).

Next, theMSC3915 responds (in Step2) with a Release message to the GANC. The GANC forwards (in Step2) this message to theUE3905 using the GA-CSR DL DIRECT TRANSFER message. TheUE3905 responds (in Step3) with the Release Complete message. This message is contained within the GA-CSR UL DIRECT TRANSFER message between UE and GANC. The GANC forwards (in Step3) the Disconnect message to the MSC. The MSC triggers (in Step4) the release of connection as described in GA-CSR connection release sub-section, above.

M. CS Handover

1. CS Handover from GERAN to GAN

a) GANC Terminates Iu Up Protocol

FIG. 40 illustrates CS handover from GERAN to GAN in some embodiments. The description of the GERAN to GAN handover procedure assumes the following: (1) the UE is on an active call on the GERAN, (2) the UE mode selection is GAN-preferred, or if GERAN/UTRAN-preferred, the RxLev from the current serving cell drops below a defined threshold. In some embodiments, this threshold can be specified as a fixed value, or provided by the GERAN BSS to the UE in dedicated mode, (3) the UE has successfully registered with a GANC, allowing the UE to obtain GAN system information, and (4) the GERAN provides information on neighboring 3G cells such that one of the cells in the 3G neighbor list matches the 3G cell information associated with the GANC, as provided in the AS-related component of the system information obtained from the GANC. As shown, theUE4005 begins to include GAN cell information in the Measurement Report message to theGERAN BSC4015. TheUE4005 reports the highest signal level for the GAN cell. This is not the actual measured signal level on GAN, rather an artificial value (e.g., RxLev=63), allowing the UE to indicate preference for the GAN.

Based on UE measurement reports and other internal algorithms, theGERAN BSC4015 decides to handover to the GAN cell. TheBSC4015 starts the handover preparation by sending (in Step2) a Handover Required message to theMSC4020, identifying thetarget 3G RNC (GANC).

TheMSC4020 requests (in Step3) thetarget GANC4010 to allocate resources for the handover using the Relocation Request message. The UE is identified by the included IMSI parameter.

Since Iu UP support mode is indicated, theGANC4010 sends (in Step4) the Iu UP INITIALISATION packet to the MSC. The MSC responds (in Step5) with the Iu UP INITIALISATION ACK packet.

TheGANC4010 builds a Handover to UTRAN Command message and sends it (in Step6) to theMSC4020 through the Relocation Request Acknowledge message. The MSC forwards (in Step7) the Handover to UTRAN Command message to theGERAN BSC4015 in the BSSMAP Handover Command message, completing the handover preparation.

Next, theGERAN BSC4015 sends (in Step8) the Intersystem to UTRAN Handover Command message, containing the Handover to UTRAN Command message, to theUE4005 to initiate handover to GAN. The UE does not switch its audio path from GERAN to GAN until handover completion (i.e., until it sends the GA-CSR HANDOVER COMPLETE message) to keep the audio interruption short.

TheUE4005 accesses (in Step9) theGANC4010 using the GA-CSR HANDOVER ACCESS message, and provides the entire Intersystem to UTRAN Handover Command message received from GERAN. TheGANC4010 sends (in Step10) the GA-CSR ACTIVATE CHANNEL message to theUE4005 including bearer path setup information such as: (1) Channel mode, (2) Multi-rate codec configuration, (3) UDP port & the IP address for the uplink RTP stream, and (4) Voice sample size.

Next, theUE4005 sends (in Step11) the GA-CSR ACTIVATE CHANNEL ACK to theGANC4010 indicating the UDP port for the downlink RTP stream. TheGANC4010 signals (in Step11) the completion of the RAB establishment to theUE4005 with the GA-CSR ACTIVATE CHANNEL COMPLETE message.

TheUE4005 transmits (in Step13) the GA-CSR HANDOVER COMPLETE message to indicate the completion of the handover procedure at its end. It switches the user from the GERAN user plane to the GAN user plane. TheGANC4010 indicates (in Step14) to theMSC4020 that it has detected the UE, using Relocation Detect message. The CN can optionally now switch the user plane from the source GERAN to the target GAN.

Bi-directional voice traffic is now (in Step15) flowing between theUE4005 andMSC4020, viaGANC4010. Thetarget GANC4010 indicates (in Step16) the handover is complete, using the Relocation Complete message. If it had not done so before, the CN now switches the user plane from source GERAN to target GAN.

The CN tears down (in Step17) the connection to the source GERAN, using Clear Command message. Finally, thesource GERAN4015 confirms (in Step18) the release of GERAN resources allocated for this call, using Clear Complete message.

b) UE terminates Iu UP Protocol

Some embodiments utilize an alternative procedure for CS handover from GERAN to GAN.FIG. 41 illustrates steps performed during GERAN to GAN in these embodiments. The description of the GERAN to GAN handover procedure assumes the following: (1) the UE is on an active call on the GERAN, (2) the UE mode selection is GAN-preferred, or if GERAN/UTRAN-preferred, the RxLev from the current serving cell drops below a defined threshold. In some embodiments, this threshold can be specified as a fixed value, or provided by the GERAN BSS to the UE in dedicated mode, (3) the UE has successfully registered with a GANC, allowing the UE to obtain GAN system information, and (4) the GERAN provides information on neighboring 3G cells such that one of the cells in the 3G neighbor list matches the 3G cell information associated with the GANC, as provided in the AS-related component of the system information obtained from the GANC.Steps1 to3 are performed the same as described forsteps1 to3 shown inFIG. 40 above and are not repeated for simplicity.

TheGANC4110 sends (in Step4) the GA-CSR ACTIVATE CHANNEL message to theUE4105 including bearer path setup information such as: (1) Channel mode, (2) Multi-rate codec configuration, (3) UDP port & the IP address for the uplink RTP stream, (4) Voice sample size, and an indication that Iu UP support mode is required. In some embodiments, theGANC4110 includes the Radio Access Bearer (RAB) parameters, and the Iu UP parameters (e.g., Iu UP mode, where support mode is used for AMR voice calls).

Since Iu UP support mode is indicated, theUE4110 sends (in Step5) the Iu UP INITIALISATION packet to the IP address and UDP port indicated in the GA-CSR ACTIVATE CHANNEL message.

TheMSC4115 responds (in Step6) with the Iu UP INITIALISATION ACK packet. TheMSC4115 sends the message to the source IP address and UDP port number of the received INITIALISATION packet. TheUE4105 sends (in Step7) the GA-CSR ACTIVATE CHANNEL ACK to theGANC4110. TheGANC4110 builds a Handover to UTRAN Command message and sends (in Step8) it to theCN4115 through the Relocation Request Acknowledge message.

TheGANC4110 signals (in Step9) the completion of the RAB establishment to theUE4105 with the GA-CSR ACTIVATE CHANNEL COMPLETE message. An end-to-end audio path now exists between theUE4105 and theMSC4115. TheMSC4115 forwards (in Step10) the Handover to UTRAN Command message to theGERAN BSC4120 in the BSSMAP Handover Command message, completing the handover preparation.

TheGERAN BSC4120 sends (in Step11) the Intersystem to UTRAN Handover Command message, containing the Handover to UTRAN Command message, to the UE to initiate handover to GAN. The UE does not switch its audio path from GERAN to GAN until handover completion (i.e., until it sends the GA-CSR HANDOVER COMPLETE message) to keep the audio interruption short.

The UE accesses theGANC4110 using (in Step12) the GA-CSR HANDOVER ACCESS message, and provides the entire Intersystem to UTRAN Handover Command message received from GERAN. TheGANC4110 indicates (in Step13) to theMSC4115 that it has detected the UE, using Relocation Detect message. TheMSC4115 can optionally now switch the user plane from the source GERAN to the target GAN. Bi-directional voice traffic is now flowing (in Step14) between the UE andMSC4115, viaGANC4110.

The UE transmits (in Step15) the GA-CSR HANDOVER COMPLETE message to indicate the completion of the handover procedure at its end. It switches the user from the GERAN user plane to the GAN user plane.

Thetarget GANC4110 indicates (in Step16) the handover is complete, using the Relocation Complete message. If it had not done so before, theMSC4115 now switches the user plane from source GERAN to target GAN.

Finally, theMSC4115 tears (in Step17) down the connection to the source GERAN, using Clear Command message. The source GERAN confirms (in Step18) the release of GERAN resources allocated for this call, using Clear Complete message.

2. CS Handover from UTRAN to GAN

a) GANC terminates Iu UP Packet

FIG. 42 illustrates CS handover from UTRAN to GAN in some embodiments. The description of the UTRAN to GAN Handover procedure assumes the following: (1) the UE is on an active call on the UTRAN, (2) the UE has been ordered by the RNC to make inter-frequency measurements (i.e., if the GAN cell has been allocated a different frequency value than is used in the UTRAN), (a) if the UE is in GAN preferred mode with an Event2A configured, the UE handles parameters associated with the Event2A in a GAN specific manner (as described in “Radio Resource Control (RRC) protocol specification”, 3GPP TS 25.331 standard, hereinafter “3GPP TS 25.331”) for the reporting of the EGAN, (b) when the UE is in GERAN/UTRAN preferred mode and an event2A has been configured for the GAN cell, the UE shall only send a measurement about the GAN cell, when this event is triggered and no UTRAN cells from the neighbor cell list of the UE satisfy the triggering condition of this Event (as described in 3GPP TS 25.331), (3) the UTRAN provides information on neighboring cells such that one of the cells in the neighbor list matches the cell associated with the GANC, as provided in the AS-related component of the system information obtained from GANC.

As shown inFIG. 42, theUE4205 begins to include information about a GAN cell in the Measurement Report message sent (in Step1) to theRNC4215. TheUE4205 reports the highest signal level for the GAN cell. This is not the actual measured signal level on the GAN, rather an artificial value allowing theUE4205 to indicate preference for the GAN.

Based on UE measurement reports and other internal algorithms, theRNC4215 decides to initiate handover to the GAN cell. TheRNC4215 starts the preparation phase of the Relocation procedure by sending (in Step2) a Relocation Required message to the MSC, identifying the target (GAN) cell.

Next, steps3 to5 shown inFIG. 42 are performed as described for steps3-5 for GERAN to GAN Handover sub-section, above. Thetarget GANC4210 acknowledges (in Step6) the handover request message, using Relocation Request Acknowledge message, indicating it can support the requested handover, and including a Physical Channel Reconfiguration message that indicates the radio channel to which the UE should be directed.

Next, theMSC4220 sends (in Step7) the Relocation Command message to theRNC4215, completing the relocation preparation. TheRNC4215 sends (in Step8) the PHYSICAL CHANNEL RECONFIGURATION message to theUE4205 to initiate handover to GAN. The UE does not switch its audio path from UTRAN to GAN until handover completion (i.e., until it sends the GA-CSR HANDOVER COMPLETE message) to keep the audio interruption short.

Next, Steps9-16 shown inFIG. 42 are performed similar to Steps9-16 for GERAN to GAN Handover described above. Next, theMSC4220 tears down (in Step17) the connection to the source RNC, using Iu Release Command. Finally, thesource RNC4215 confirms (in Step18) the release of UTRAN resources allocated for this call, using Iu Release Complete.

b) UE Terminates Iu UP Packet

Some embodiments utilize an alternative procedure for CS handover from UTRAN to GAN.FIG. 43 illustrates steps performed during UTRAN to GAN in these embodiments. As shown, the UE begins to include (in Step1) information about a GAN cell in the Measurement Report message sent to theRNC4320. The UE reports the highest signal level for the GAN cell. This is not the actual measured signal level on the GAN, rather an artificial value allowing the UE to indicate preference for the GAN.

Based on UE measurement reports and other internal algorithms, theRNC4320 decides to initiate handover to the GAN cell. TheRNC4320 starts the preparation phase of the Relocation procedure by sending (in Step2) a Relocation Required message to theMSC4315, identifying the target GAN cell.

TheMSC4315 requests (in Step3) thetarget GANC4310 to allocate resources for the handover using the Relocation Request message. TheUE4305 is identified by the included IMSI parameter.

TheGANC4310 sends (in Step4) the GA-CSR ACTIVATE CHANNEL message to theUE4305 including bearer path setup information received in the Relocation Request message, such as: (1) UDP port & the IP address for the uplink RTP stream, (2) Radio Access Bearer (RAB) parameters, and (3) Iu UP parameters (e.g., Iu UP mode, where support mode is used for AMR voice calls).

Since Iu UP support mode is indicated, theUE4305 sends (in Step5) the Iu UP INITIALISATION packet to the IP address and UDP port indicated in the GA-CSR ACTIVATE CHANNEL message. This message is routed to the core network4315 (e.g., the R4 media gateway).

TheMSC4315 responds (in Step6) with the Iu UP INITIALISATION ACK packet. TheMSC4315 sends the message to the source IP address and UDP port number of the received INITIALISATION packet. TheUE4305 sends (in Step7) the GA-CSR ACTIVATE CHANNEL ACK to theGANC4310.

Thetarget GANC4310 acknowledges (in Step8) the handover request message, using Relocation Request Acknowledge message, indicating it can support the requested handover, and including a Physical Channel Reconfiguration message that indicates the radio channel to which theUE4305 should be directed.

TheGANC4310 signals (in Step9) the completion of the RAB establishment to theUE4305 with the GA-CSR ACTIVATE CHANNEL COMPLETE message. An end-to-end audio path now exists between theUE4305 and theMSC4315. TheMSC4315 sends (in Step10) the Relocation Command message to theRNC4320, completing the relocation preparation.

TheRNC4320 sends (in Step11) the PHYSICAL CHANNEL RECONFIGURATION message to the UE to initiate handover to GAN. The UE does not switch its audio path from UTRAN to GAN until handover completion (i.e., until it sends the GA-CSR HANDOVER COMPLETE message) to keep the audio interruption short. The UE accesses (in Step12) theGANC4310 using the GA-CSR HANDOVER ACCESS message, and provides the entire PHYSICAL CHANNEL RECONFIGURATION message received fromRNC4320.

TheGANC4310 indicates (in Step13) to theMSC4315 that it has detected the UE, using Relocation Detect message. TheMSC4315 can optionally now switch the user plane from thesource RNC4320 to thetarget GANC4310. Bi-directional voice traffic is now flowing (in Step14) between the UE andMSC4315, viaGANC4310.

The UE transmits (in Step15) the GA-CSR HANDOVER COMPLETE to indicate the completion of the handover procedure from its perspective. It switches the user from the UTRAN user plane to the GAN user plane. Thetarget GANC4310 indicates (in Step16) the handover is complete, using the Relocation Complete message. If it has not done so before, theCN4315 now switches the user plane fromsource RNC4320 to targetGANC4310.

Finally, theMSC4315 tears (in Step17) down the connection to thesource RNC4320, using Iu Release Command. Thesource RNC4320 confirms (in Step18) the release of UTRAN resources allocated for this call, using Iu Release Complete.

3. CS Handover from GAN to GERAN

FIG. 44 illustrates the procedure to handover from GAN to GERAN in some embodiments. The procedure description in this sub-clause assumes the following: (1) the UE is on an active call in GAN Iu-mode, and (2) the GERAN becomes available and (a) the UE mode selection is GERAN/UTRAN-preferred, or (b) the UE mode selection is GAN-preferred and the UE begins to leave GAN coverage, based on its local measurements, received RTCP reports, as well as any uplink quality indications received from the GANC. The handover from GAN to GERAN procedure is always triggered by the UE. As shown inFIG. 44, the following steps are performed during handover from GAN to GERAN.

TheGANC4410 may send (in Step1) a GA-CSR UPLINK QUALITY INDICATION when there is a problem with the uplink quality for the ongoing call. Uplink Quality Indication is information sent by the GANC to the UE indicating the crossing of an uplink quality threshold in the uplink direction. Whenever the UE receives an indication of bad quality, it should start the handover procedure, as described in the next step. Alternatively, UE can use its local measurements or received RTCP reports, to decide to initiate the handover procedure.

As shown, theUE4405 sends (in Step2) the GA-CSR HANDOVER INFORMATION message to theGANC4410 indicating the Channel Mode and a list of target GERAN cells, identified by CGI, in order of preference (e.g. ranked by C1 path loss parameter) for handover, and includes the received signal strength for each identified GERAN cell. This list is the most recent information available from the GSM RR subsystem. In addition, the GA-CSR HANDOVER INFORMATION message may include a list of target UTRAN cells ranked in order of preference for handover, and the received signal strength for each identified UTRAN cell.

If the Serving GANC selects a target GERAN cell, the handover to GERAN procedure is performed. TheServing GANC4410 starts the handover preparation by signaling (in Step3) to theMSC4420 the need for handover, using Relocation Required, and including the GERAN cell list provided by the UE. The GANC may include only a subset of the cell list provided by the UE.

TheMSC4420 then selects a target GERAN cell and requests it (in Step4) to allocate the necessary resources, using Handover Request. Thetarget GERAN BSC4415 builds a Handover Command message providing information on the channel allocated and sends it (in Step5) to theMSC4420 through the Handover Request Acknowledge message.

TheMSC4420 signals (in Step6) theGANC4410 to handover theUE4405 to the GERAN, using Relocation Command message, ending the handover preparation phase. The GANC transmits (in Step7) the GA-CSR HANDOVER COMMAND to the UE including the details sent by the GERAN on the target resource allocation.

Next, theUE4405 transmits (in Step8) the “Um: Handover Access” message containing the handover reference element to allow thetarget GERAN BSC4415 to correlate this handover access with the Handover Command message transmitted earlier to the MSC in response to the Handover Required. Thetarget GERAN BSC4415 confirms (in Step9) the detection of the handover to theMSC4420, using the Handover Detect message.

TheMSC4420 may at this point switch (in Step10) the user plane to the target BSS. TheGERAN BSC4415 provides (in Step11) Physical Information to the UE (i.e., Timing Advance) to allow the UE to synchronize with the GERAN. TheUE4405 signals (in Step12) to theGERAN BSC4415 that the handover is completed, using Handover Complete.

TheGERAN BSC4415 confirms (in Step13) to theMSC4420 the completion of the handover, via Handover Complete message. TheMSC4420 may use the target CGI used in the Handover procedure for charging purposes.

Bi-directional voice traffic is now flowing (in Step14) between theUE4405 andMSC4420, via theGERAN BSC4415. On receiving the confirmation of the completion of the handover, theMSC4420 indicates (in Step15) to the GANC to release any resources allocated to the UE, via the Iu Release Command.

Next,GANC4415 commands (in Step16) theUE4405 to release resources, using the GA-CSR RELEASE message. TheGANC4410 confirms (in Step17) resource release toMSC4420 using the Iu Release Complete message.

TheUE4405 confirms (in Step18) resource release to theGANC4410 using the GA-CSR RELEASE COMPLETE message. TheUE4405 may finally deregister (in Step19) from the GANC, using GA-RC DEREGISTER message.

4. CS Handover from GAN to UTRAN

FIG. 45 illustrates the procedure to handover from GAN to UTRAN in some embodiments. The procedure description assumes the following: (1) the UE is on an active call on the GAN, (2) the UE is capable of operating in all of the GAN, GERAN and UTRAN modes, (3) the UTRAN becomes available and (a) the UE is in GERAN/UTRAN-preferred mode, or (b) the UE mode selection is GAN preferred and begins to leave GAN coverage, based on its local measurements, received RTCP reports, as well as any uplink quality indications received from the GANC. The handover from GAN procedure is always triggered by the UE. As shown inFIG. 45, the following steps are performed during handover from GAN to UTRAN.

TheGANC4510 may send (in Step1) a GA-CSR UPLINK QUALITY INDICATION if there is a problem with the uplink quality for the ongoing call. Uplink Quality Indication is information sent by theGANC4510 to theUE4505 indicating the crossing of an uplink quality threshold in the uplink direction. Whenever theUE4505 receives an indication of bad quality, it should start the handover procedure, as described in the next step. Alternatively, UE can use its local measurements or received RTCP reports, to decide to initiate the handover procedure.

Next, theUE4505 sends (in Step2) the GA-CSR HANDOVER INFORMATION message to the Serving GANC indicating the Channel Mode and a list of candidate target UTRAN and GERAN cells, in order of preference for handover, and includes the received signal strength for each identified cell. The UTRAN cells are identified by the PLMN ID, the LAC and the 3G Cell identity (defined in 3GPP TS 25.331).

If theServing GANC4510 selects UTRAN as the target RAT, the handover to UTRAN procedure is performed. TheServing GANC4510 starts the handover preparation by signaling (in Step3) to theMSC4520 the need for handover, using Relocation Required and including the UTRAN cell list provided by theUE4505. TheGANC4510 may include only a subset of the cell list provided by theUE4505.

TheMSC4520 starts the handover procedure towards thetarget RNC4515 identified by the Serving GANC. TheMSC4520 requests (in Step4) from thetarget RNC4515 to allocate the necessary resources using Relocation Request. Thetarget RNC4515 builds a Physical Channel Reconfiguration message providing information on the allocated UTRAN resources and sends it (in Step5) to theMSC4520 through the Relocation Request Acknowledge message.

Next, theMSC4520 signals (in Step6) theServing GANC4510 to handover the UE to the UTRAN, using Relocation Command message (which includes the Physical Channel Reconfiguration message), ending the handover preparation phase.

TheServing GANC4510 transmits (in Step7) the GA-CSR HANDOVER COMMAND to the UE including the details sent by the UTRAN on the target resource allocation.Target RNS4515 achieves (in Step8) uplink synchronization on the Uu interface.

Thetarget RNC4515 confirms (in Step9) the detection of the handover to the MSC, using the Relocation Detect message. TheMSC4520 may at this point switch (in Step10) the user plane to thetarget RNS4515.

Next, theUE4505 signals (in Step11) to theUTRAN RNC4515 that the handover is completed, using Handover to UTRAN Complete. TheUTRAN RNC4515 confirms (in Step12) to theMSC4520 the completion of the handover, via Relocation Complete message. If the user plane has not been switched instep10, theMSC4520 switches the user plane to the target RNS.

Bi-directional voice traffic is now flowing (in Step13) between theUE4505 andMSC4520, via theUTRAN RNC4515. On receiving the confirmation of the completion of the handover, theMSC4520 indicates (in Step14) to theServing GANC4510 to release any resources allocated to the UE, via the Iu Release Command.

TheServing GANC4510 then commands (in Step15) theUE4505 to release resources, using the GA-CSR RELEASE message. TheServing GANC4510 confirms (in Step16) resource release toMSC4520 using the Iu Release Complete message.

TheUE4505 confirms (in Step17) resource release to theServing GANC4510 using the GA-CSR RELEASE COMPLETE message. TheUE4505 may finally deregister (in Step18) from theServing GANC4510, using GA-RC DEREGISTER message.

N. GA-PSR Connection Handling

The Iu-mode GA-PSR connection is a logical connection between the UE and the GANC for the PS domain. A GA-PSR connection is established when the upper layers in the UE request the establishment of a PS domain signaling connection and the UE is in GA-PSR-IDLE state; i.e., no GA-PSR connection exists. When a successful response is received from the network, GA-PSR replies to the upper layer that the PS domain signaling connection is established and the UE has entered the equivalent of the RRC connected mode (i.e., the GA-PSR-CONNECTED state).

1. GA-PSR Connection Establishment

FIG. 46 illustrates successful and unsuccessful establishment of the GA-PSR Connection in some embodiments. As shown, theUE4605 initiates GA-PSR connection establishment by sending (in Step1) the GA-PSR REQUEST message to theGANC4610. This message contains the Establishment Cause indicating the reason for GA-PSR connection establishment. When theGANC4610 determines that the GA-PSR connection request can be accepted, theGANC4610 signals the acceptance of the connection request to theUE4605 by sending (in Step2) the GA-PSR REQUEST ACCEPT and the UE enters the GA-PSR-CONNECTED state. Alternatively, when theGANC4610 determines that the GA-PSR connection request has to be rejected, theGANC4610 sends (in Step3) a GA-PSR REQUEST REJECT to the UE ZC05 indicating the reject cause, completing the procedure.

2. GA-PSR Connection Release

FIG. 47 illustrates release of the logical GA-PSR connection between the UE and the GANC in some embodiments. The following steps are performed during the release. As shown, theSGSN4715 indicates to theGANC4710 to release the PS resources allocated to the UE by sending (in Step1) the RANAP Iu Release Command message to theGANC4710.

Next, theGANC4710 confirms (in Step2) resource release toSGSN4715 using the Iu Release Complete message. Next, theGANC4710 commands (in Step3) theUE4705 to release resources, using the GA-PSR RELEASE message. Finally, theUE4705 confirms (in Step4) resource release to theGANC4710 using the GA-PSR RELEASE COMPLETE message and the GA-PSR state in the UE changes to GA-PSR-IDLE.

O. PS Security Mode Control

FIG. 48 illustrates the message flow for PS security mode control in some embodiments. As shown,SGSN4815 sends (in Step1) the RANAP Security Mode Command message toGANC4810. This message contains the integrity key (IK) and allowed algorithms, and optionally the encryption key (CK) and allowed algorithms.

Next, theGANC4810 sends (in Step2) the GA-PSR SECURITY MODE COMMAND message to theUE4805. This message indicates the integrity protection and encryption settings (i.e., that are applicable after relocation to UTRAN), and a random number. The UE stores the information for possible future use after a relocation to UTRAN.

Next, theUE4805 computes a message authentication code (MAC) based on the random number, the UE IMSI and the integrity key calculated by the UE. TheUE4805 then sends (in Step3) the GA-PSR SECURITY MODE COMPLETE message to theGANC4810 to signal its selected algorithm and the computed MAC.

TheGANC4810 then verifies the MAC using the random number, the UE IMSI and the integrity key provided by the SGSN instep1. When the GANC verifies the MAC to be correct it sends (in Step4) the Security Mode Complete message to theSGSN4815. The MAC proves that the identity that is authenticated to the GANC is the same as the identity authenticated to the core network.

P. PS NAS Signaling Procedures

After GA-PSR connection establishment, NAS signaling may be transfer from SGSN-to-UE and from UE-to-SGSN.

1. SGSN-to-UE NAS Signaling

FIG. 49 illustrates SGSN-to-UE PS NAS signaling in some embodiments. As shown, for SGSN-to-UE NAS signaling, theSGSN4915 sends (in Step1) a NAS PDU to the GANC via the RANAP Direct Transfer message. TheGANC4910 encapsulates the NAS PDU within a GA-PSR DL DIRECT TRANSFER message and forwards (in Step2) the message to theUE4905 via the existing TCP connection.

2. UE-to-SGSN NAS Signaling

FIG. 50 illustrates UE-to-SGSN NAS signaling in some embodiments. As shown, theUE5005 receives a request from the NAS layer to transfer an uplink NAS PDU. Assuming the required signaling connection already exists, theUE5005 encapsulates the NAS PDU within a GA-PSR UL DIRECT TRANSFER message and sends (in Step1) the message to theGANC5010. TheGANC5010 relays (in Step2) the received message to theSGSN5015 that is currently serving the UE via the RANAP Direct Transfer message.

Q. GA-PSR Packet Transport Channel Management Procedures

The GA-PSR Packet Transport Channel (GA-PSR PTC) provides the association between the UE and the network for the transport of GPRS user data over the Up interface (i.e., via the GAN in Iu-mode). The PTC uses the GTP-U protocol running over UDP transport. The endpoint addresses of the PTC are identified by the IP addresses and UDP ports assigned to the PTC in the UE and network during the PTC activation procedure. The UDP port number for GTP-U is as defined in “UTRAN Iu interface data transport & transport signalling”, 3GPP TS 25.414 standard, hereinafter “3GPP TS 25.414”.

Multiple PTC instances between a UE and the network may be activated at the same time, using the same endpoint addresses. Each PTC instance is assigned unique GTP-U Tunnel Endpoint IDs (one on the UE and one on the network) during the activation procedure. The UE and GANC manage the activation and deactivation of the PTC instances based on the requests for data transfer and the configurable PTC Timer.

1. States of the GA-PSR Packet Transport Channel

The UE in the GA-PSR-CONNECTED state can be in one of two PTC substates: PTC-STANDBY or PTC-ACTIVE. The PTC-STANDBY substate is the initial/default PTC substate of the UE when in the GA-PSR-CONNECTED state in GAN mode. The UE is not able to send or receive GPRS user data to or from the network. The UE needs to activate the PTC before sending any GPRS user data. When the UE successfully establishes a PTC, the UE transitions to the PTC-ACTIVE substate.

In PTC-ACTIVE substate, the UE is in the GA-PSR-CONNECTED state and the PTC is active between the UE and the network and the UE is able to send and receive GPRS user data to and from the network. Several events can trigger the GA-PSR PTC activation on the UE side. These events include the UE initiates the uplink user data transfer or the GANC initiates PTC activation; i.e., the UE receives a GA-PSR-ACTIVATE-PTC-REQUEST message from the GANC.

On successful PTC activation and in parallel with transition to the PTC-ACTIVE substate, the UE starts the PTC Timer. When the PTC Timer expires, the UE sends a message to the GANC to initiate PTC deactivation. On successful PTC deactivation, the UE transitions to PTC-STANDBY substate.

At any time while in the GA-PSR-CONNECTED state and the PTC-ACTIVE substate, the UE may receive the GA-PSR RELEASE message. In addition to requesting release of the GA-PSR session, this is interpreted by the UE as an implicit PTC deactivate command.

At any time while in GAN mode, if the serving RR entity is switched to GSM-RR/UTRAN-RRC, the GA-PSR is disconnected from the GPRS SAPs and the UE enters GERAN/UTRAN mode. Simultaneously, the UE will release the associated PTC regardless of the PTC Timer status.

The UE GA-PSR entity maintains one PTC for each active PDP context. The PTC Timer is restarted whenever any uplink user data packet is sent or downlink user data packet is received related to the PDP context. The PTC Timer value is provided to the UE as part of the GAN Registration procedure (i.e., in the GA-RC REGISTER ACCEPT message).

2. PTC Initial Activation

FIG. 51 depicts the Packet Transport Channel initial activation procedure, assuming the UE is in the GA-PSR-IDLE state. As shown, the following steps are performed. The GA-PSR Connection Establishment procedure is performed (in Step1) as described in GA-PSR connection establishment sub-section, above. TheUE5105 transitions to the GA-PSR-CONNECTED state and the PTC-STANDBY substate. Next, additional PS signaling procedures are performed (in Step2). Examples of these signaling procedures are illustrated in PDP Context Activation and Network Requested PDP Context Activation sub-sections, below.

Next, theSGSN5115 initiates (in Step3) the RAB Assignment procedure and includes the RAB-ID, the CN Transport Layer Address (IP address) and the CN Iu Transport Association (GTP-U Terminal Endpoint Identifier, TEID) for user data. TheGANC5110 sends (in Step4) the GA-PSR ACTIVATE PTC REQUEST message to the UE to request activation of the Packet Transport Channel. The message includes the RAB-ID, a TEID that the GANC assigns to the UE, and the GANC IP Address and GANC TEID. If the GANC is configured to allow the UE to send PTC packets (i.e., GTP-U messages) directly to the SGSN (i.e., the configuration illustrated inFIG. 17), the GANC sets the GANC IP Address to the CN IP Address and the GANC TEID to the CN TEID; otherwise, the GANC assigns a local address as the GANC IP address and a GANC-allocated TEID as the GANC TEID and sends this information to the UE (i.e., the configuration illustrated inFIG. 18). TheUE5105 acknowledges (in Step5) the PTC activation.

TheGANC5110 sends (in Step6) the RAB Assignment Response message to theSGSN5115 to complete the RAB Assignment procedure. If the GANC is configured to allow theSGSN5115 to send GTP-U messages directly to the UE5105 (i.e., the configuration illustrated inFIG. 17), theGANC5110 sets the RAN IP Address to the UE's IP Address and the RAN TEID to the TEID assigned to the UE by the GANC; otherwise, the GANC assigns a local address as the RAN IP address and a GANC-allocated TEID as the RAN TEID and sends this information to the SGSN (i.e., the configuration illustrated inFIG. 18).

Next, theGANC5110 signals (in Step7) the completion of the RAB establishment to theUE5105 with the GA-PSR ACTIVATE PTC COMPLETE message. On receipt of the message, the UE transitions to the PTC-ACTIVE substate and starts the PTC Timer. Next, additional PS signaling procedures are performed (in Step8). Examples of these PS signaling are illustrated in PDP Context Activation and Network Requested PDP Context Activation sub-sections, below. TheUE5105 initiates (in Step9) uplink user data transfer via the established PTC and theSGSN5115 may use the same transport channel to send downlink user data packets.

3. PTC Data Transfer

FIG. 52 illustrates the transfer of GPRS user data packets via the GAN Packet Transport Channel. This scenario assumes that user data is carried transparently between the UE and core network (i.e., the configuration illustrated inFIG. 17). As shown, the following steps are performed.

When required, the GAN PTC is established (in Step1) as specified in PCT Intial Activation sub-section, above. Upon the GA-PSR PTC establishment, theUE5205 enters the PTC-ACTIVE substate and starts the PTC Timer. Next, theUE5205 initiates (in Step2) the transfer of an uplink user data packet using the standard GTP-U protocol as specified in “GPRS Tunnelling Protocol (GTP) across the Gn and Gp interface”, 3GPP TS 29.060 standard, hereinafter “3GPP TS 29.060” and restarts the PTC Timer.

Next, theSGSN5215 transfers (in Step3) downlink user data packet utilizing the same PTC associated with the specific PDP context. Downlink user data packets are transferred using the standard GTP-U protocol as specified in 3GPP TS 29.060. Upon receiving the downlink data packet, the UE restarts the associated PTC Timer. Additional uplink and downlink user data packets are transferred (in Step4) via the same PTC as described in

steps

2 and3, respectively. After each transmission/reception, the UE restarts the PTC Timer. If the configuration illustrated inFIG. 18 is used, then the uplink GTP-U packets are sent from UE to GANC, then relayed from GANC to SGSN; likewise, downlink GTP-U packets are sent from SGSN to GANC, then relayed from GANC to UE.

4. MS Initiated PTC Deactivation

FIG. 53 depicts the scenario when the UE deactivates the Packet Transport Channel after the PTC Timer expires. The UE is in the GA-PSR-CONNECTED state and the PTC-ACTIVE substate. As shown, the following steps are performed.

The PTC Timer associated with one of the active packet transport channels expires (in Step1). TheUE5305 sends (in Step2) the GA-PSR DEACTIVATE PTC REQUEST message to theGANC5310, including the RAB-ID to identify the PTC and indicating the normal release as a cause for deactivation. Alternatively, the UE may indicate PTC timer expiry as the cause for deactivation.

Next, theGANC5310 sends (in Step3) a RAB Release Request message to theSGSN5315 to request the release of the associated RAB. TheSGSN5315 responds (in Step4) with the RAB Assignment Request indicating release.

TheGANC5310 responds (in Step5) to theUE5305 with a GA-PSR DEACTIVATE PTC ACK message to acknowledge successful deactivation. TheUE5305 transitions to the PTC-STANDBY substate. TheGANC5310 sends (in Step6) the RAB Assignment Response message to notify theSGSN5315 that the RAB Release procedure is complete.

5. MS Initiated PTC Re-Activation

FIG. 54 depicts the scenario when in some embodiments the UE initiates re-activation of the Packet Transport Channel while in the GA-PSR-CONNECTED and PMM-CONNECTED states; e.g., a PS signaling connection and active PDP context exists between the UE and CN but the PTC was previously deactivated by the UE due to PTC Timer expiry. As shown, the following steps are performed. The UE is in the GA-PSR-CONNECTED state and the PTC-STANDBY substate. The UE is in the PMM-CONNECTED state (i.e., a PS signaling connection and an active PDP context exists).

When theUE5405 has a PDU to send, theUE5405 sends (in Step1) the Service Request message (with Service type value “Data”) to theGANC5410 in the GA-PSR UL DIRECT TRANSFER message. Next, theGANC5410 forwards (in Step2) the Service Request over the existing signaling connection to theSGSN5415 using the RANAP Direct Transfer message.

TheSGSN5415 may optionally initiate (in Step3) the Security Mode Control procedure described in Security Mode Control sub-section, above. TheSGSN5415 sends (in Step4) a Service Accept message to theGANC5410. TheGANC5410 forwards (in Step5) the message to the UE.

Next, theUE5405,GANC5410 andSGSN5415 establish (in Step6) the GA-PSR Packet Transport Channel (PTC) as described in steps3-7 in PTC Initial Activation Sub-section, above. The UE transitions to the PTC-ACTIVE substate and starts the PTC Timer. Finally, theUE5405 sends (in Step7) the uplink PDU. Additional data transfer may also take place.

6. Network Initiated PTC De-Activation

FIG. 55 depicts the scenario when the network initiates de-activation of the Packet Transport Channel in some embodiments. The UE is in the GA-PSR-CONNECTED state and the PTC-ACTIVE substate. As shown, the following steps are performed.

Optionally, theGANC5510 may initiate the PTC de-activation procedure; e.g., as a result of an error handling procedure. If so, theGANC5510 sends (in Step1) the RAB Release Request message to theSGSN5515.

TheSGSN5515 sends (in Step2) a RAB Assignment Request to request the release of the associated RAB. The release request may include one or more RABs. Next, theGANC5510 requests deactivation of the associated GA-PSR PTC by sending (in Step3) the GA-PSR DEACTIVATE PTC REQUEST message to theUE5505.

TheUE5505 transitions to the PTC-STANDBY substate, stops the PTC Timer and sends (in Step4) the acknowledgment back to the GANC.

Steps

3 and4 are repeated for each additional RAB/PTC that needs to be released. Finally, theGANC5510 notifies (in Step5) theSGSN5515 that the release was successful.

7. Network Initiated PTC Re-Activation

FIG. 56 depicts the scenario in some embodiments when the network initiates re-activation of the Packet Transport Channel while the UE is in the GA-PSR-CONNECTED and PMM-CONNECTED states; e.g., a PS signaling connection and active PDP context exists between the UE and CN but the PTC was previously deactivated. The UE is in the GA-PSR-CONNECTED state and the PTC-STANDBY substate. The UE is in the PMM-CONNECTED state (i.e., a PS signaling connection and an active PDP context exists). As shown, the following steps are performed.

When theSGSN5615 has a PDU to send to theUE5605, theSGSN5615 may optionally initiate (in Step1) the Security Mode Control procedure described in Security Mode Control sub-section, above. TheUE5605,GANC5610 andSGSN5615 establish (in Step2) the GA-PSR Packet Transport Channel (PTC) as described in steps3-7 in PTC Initial Activation sub-section, above. The UE transitions to the PTC-ACTIVE substate and starts the PTC Timer. TheSGSN5615 then sends (in Step3) the downlink PDU. Additional data transfer may also take place.

8. Implicit PTC De-Activation Due to UE De-Registration

As part of the GAN de-registration procedure, the GANC needs to release all resources allocated to the UE. GAN de-registration may be initiated either explicitly by the UE or implicitly by the GANC if the loss of the signaling connection is detected (as described in De-Registration sub-section, above).FIG. 57 illustrates implicit PTC deactivation procedure in some embodiments. Initially, one or more GA-PSR PTCs associated with a UE are in the PTC-ACTIVE state. As shown, the following steps are performed.

The GAN de-registration procedure is initiated (in Step1) for theUE5705 either by theUE5705 orGANC5710. Optionally, any outstanding resources associated with the CS Domain are released (in Step2).

TheGANC5710 initiates (in Step3) the Iu release procedure to release the corresponding RABs. TheSGSN5715 responds (in Step4) with Iu Release Command.

Upon receiving the Iu Release Command, theGANC5710 locally deactivates (in Step6) all associated PTCs and responds (in Step6) to theSGSN5715 with an Iu Release Complete message.

R. PDP Context Activation

FIG. 58 illustrates the successful UE-initiated PDP Context Activation procedure in some embodiments, assuming the UE is in GA-PSR-IDLE state. As shown, the following steps are performed.

The GA-PSR Connection Establishment procedure is performed (in Step1) as described in GA-PSR Connection Establishment Sub-section, above. TheGANC5810 establishes an SCCP connection to the SGSN and forwards (in Step2) the Service Request message (with Service type value “Signaling”) to theSGSN5815 using the RANAP Initial UE Message. Subsequent NAS messages between the UE and core network will be sent between GANC and SGSN using the RANAP Direct Transfer message.

TheSGSN5815 may optionally authenticate (in Step3) the UE using standard UTRAN authentication procedures. TheSGSN5815 may optionally initiate (in Step4) the Security Mode Control procedure described in Security Mode Control Sub-section, above. TheSGSN5815 responds (in Step5) with a Service Accept message. TheGANC5810 forwards (in Step5) the message to theUE5805.

TheUE5805 then sends (in Step6) the Activate PDP Context Request message providing details on the PDP context to theSGSN5815. This message is contained within the GA-PSR UL DIRECT TRANSFER between theUE5805 and theGANC5810. TheGANC5810 forwards (in Step6) the Activate PDP Context Request message to theSGSN5815.

Next, theUE5805,GANC5810, andSGSN5815 establish (in Step7) the GA-PSR Packet Transport Channel (PTC) as described in steps3-7 in PTC Initial Activation, above. TheSGSN5815 indicates (in Step8) the PDP context establishment is complete using the Activate PDP Context Accept message to the GANC. GANC forwards this message to the UE in the GA-PSR DL DIRECT TRANSFER message. Finally, theUE5805 andCN5815 exchange (in Step9) user data transfer via the established PTC.

S. Network Requested PDP Context Activation

FIG. 59 illustrates the successful Network-Requested PDP Context Activation procedure in some embodiments, assuming the UE is in GA-PSR-IDLE state. Initially, the SGSN received downlink user data to transfer to the UE and the associated RAB is not established. The UE is in PMM-IDLE state. As shown, theSGSN5915 sends (in Step1) the RANAP Paging message to theUE5905 via theGANC5910 to locate the user. The paging request indicates paging for PS Domain signaling.

TheGANC5910 forwards (in Step2) the paging information to theUE5905 in the GA-PSR PAGING REQUEST message. The GA-PSR Connection Establishment procedure is performed (in Step3) as described in GA-PSR Connection Establishment Sub-section, above. Alternatively, rather than using the GA-PSR Connection Establishment procedure, theUE5905 may send the GA-PSR PAGING RESPONSE message (in Step3) and then transition to the GA-PSR CONNECTED state.

TheGANC5910 establishes an SCCP connection to the SGSN and forwards (in Step4) the Service Request message (with Service type value “Paging response”) to theSGSN5915 using the RANAP Initial UE Message. Subsequent NAS messages between theUE5905 andcore network5915 will be sent betweenGANC5910 andSGSN5915 using the RANAP Direct Transfer message.

TheSGSN5915 may optionally authenticate (in Step5) theUE5905 using standard UTRAN authentication procedures. TheSGSN5915 may optionally initiate (in Step6) the Security Mode Control procedure described in Security Mode Control Sub-section, above.

Next, theSGSN5915 sends (in Step7) the Request PDP Context Activation message to theGANC5910. TheGANC5910 forwards (in Step7) this message to theUE5905 in the GA-PSR DL DIRECT TRANSFER message. TheUE5905 sends (in Step8) the Activate PDP Context Request message providing details on the PDP context to theSGSN5915. This message is contained within the GA-PSR UL DIRECT TRANSFER between the UE and the GANC. TheGANC5910 forwards (in Step8) the Activate PDP Context Request message to theSGSN5915.

TheUE5905,GANC5910, andSGSN5915 establish (in Step9) the GA-PSR Packet Transport Channel (PTC) as described in steps3-7 in PTC Initial Activation Sub-section, above. TheSGSN5915 indicates (in Step10) the PDP context establishment is complete using the Activate PDP Context Accept message to the GANC. GANC forwards this message to the UE in the GA-PSR DL DIRECT TRANSFER message. Finally, theUE5905 andSGSN5915 exchange (in Step11) user data transfer via the established PTC.

T. SRNS Relocation Between UTRAN and GAN

The SRNS Relocation procedure is performed to move one or more PS sessions between Iu mode GAN and UTRAN. It relocates the Iu-ps connection point at the GAN/UTRAN (in all cases) and at the SGSN (for inter-SGSN Relocation case only).

Support for the Iur interface between UTRAN and GAN is not described in this document. Therefore, only the Combined Hard Handover and SRNS Relocation is applicable for GAN—UTRAN SRNS Relocation. Consequently, only the “UE Involved” Relocation Type is supported.

1. SRNS Relocation from UTRAN to GAN

A) Preparation Phase

FIG. 60 illustrates the UTRAN to GAN SRNS relocation preparation phase in some embodiments. As shown, the following steps are performed.

TheUE6005 has one or more active PDP Contexts with active RABs in the UTRAN. Next, theUE6005 detects aGAN6015, performs (in Step2) the Registration procedures and enters GA-RC-REGISTERED state with valid GAN cell identity information.

The Measurement Control message (in Step3) from theRNC6010 toUE6005 includes this GAN's cell identity. The UE begins to include the GAN cell information in the Measurement Report sent (inStep3a) to the RNC. In that message, it sets the GAN cell's signal strength indicator to the highest possible value.

Next, theRNC6010 decides to initiate a Combined Hard Handover and SRNS Relocation procedure. This decision is made based on the measurement reports and vendor/operator specific criteria. Upon deciding to initiate the Relocation, theRNC6010 sends (in Step4) Relocation Required to the SGSN.

TheSGSN6020 determines the target cell is the GANC, based on the contents of Relocation Required. TheSGSN6020 then sends (in Step5) the Relocation Request to theGANC6015.

Upon receiving Relocation Request message, theGANC6015 will setup (in Step6) Packet Transport Channel(s) as described in

steps

4,5 and7 in PTC Initial Activation Sub-section, above as needed with appropriate attributes, as defined in the message. TheGANC6015 will then send (inStep6a) a Relocation Request Acknowledge to the SGSN.

b) Execution Phase

FIG. 61 illustrates UTRAN to GAN SRNS Relocation Execution Phase in some embodiments. As shown, the following steps are performed.

Upon receiving the positive acknowledgement from theGANC6115 to serve theUE6105, theSGSN6120 initiates the Execution Phase by sending (in Step1) the Relocation Command to theRNC6110. TheRNC6110 instructs theUE6105 by sending (inStep2a) the Physical Channel Reconfiguration message to initiate the physical layer switch to move to the GAN.

When the QoS attributes of any of the active RABs require lossless in-sequence SDU Delivery (lossless PDCP), then theRNC6110 starts forwarding (inStep2b) GTP PDUs to theGANC6115 while still transmitting them in the downlink direction to theUE6105. This forwarding is routed via the Iu_PS interface. The GANC may buffer, transmit in the downlink, or discard these forwarded GTP PDUs, depending on the QoS profile, network conditions, and whether it supports Lossless Relocation. Specific implementation is vendor and/or operator specific. In addition, the GANC may delay the start of the downlink transmission untilStep5 below to synchronize the GTP-U sequence numbers.

The RNC sends (inSteps2cand3a) the Forward SRNS Context message to the GAN via the SGSN. In this message, the next-expected sequence number of uplink and downlink GTP-U packets are indicated to the GANC by the old SRNC. If the QoS attributes require and GANC supports Lossless Relocation, then these sequence numbers are used to ensure in-sequence delivery of GTP PDUs.

Immediately after receiving the Physical Channel Reconfiguration message, theUE6105 sends (inStep3b) GA-PSR-HANDOVER-COMPLETE message to theGANC6115. Upon receiving this message and the Forward SRNS Context message sent from the SGSN6120 (inStep3a), theGANC6115 becomes the Serving RNC.

Immediately upon receiving the GA-PSR-HANDOVER-COMPLETE message from the UE, theGANC6115 sends (in Step4) the Relocation Detect message to theSGSN6120. When the UE supports Lossless Relocation and one or more RABs QoS attribute requires it, the UE initiates (in Step5) a GTP-U sequence number exchange procedure with the GANC over the newly established PTC. When theGANC6115 supports Lossless Relocation and one or more RABs QoS attribute requires it, it may also initiate a GTP-U sequence number exchange procedure, if the procedure had not been already initiated by the UE.

Upon completion of the GTP-U sequence number exchange procedure, theGANC6115 sends (in Step6) Relocation Complete message to the SGSN. If the GTP-U sequence number exchange is skipped (either due to lack of support in UE and/or GAN or QoS attributes did not require it), then the Relocation Complete is sent right after the Relocation Detect message. The active RABs and PDP contexts are now moved to between UE, GANC and SGSN. TheSGSN6120 then releases (in Step7) the Iu_PS connection with theold RNC6110. When the Routing Area of the GANC cell (as indicated by the GANC to the UE) is different from that under the old RNC, then theUE6105 performs (in step8) Routing Area Update procedure.

2. SRNS Relocation from GAN to UTRAN

a) Preparation Phase

FIG. 62 illustrates GAN to UTRAN SRNS Relocation Preparation Phase in some embodiments. As shown, the followings steps are performed.

TheUE6205 is (in Step1) in active packet flow exchange with active PDP Context(s) and PTCs in the GAN. TheGANC6215 may send (in Step2) a GA-PSR UPLINK QUALITY INDICATION if there is a problem with the uplink quality for the ongoing session. Uplink Quality Indication is information sent by theGANC6215 to theUE6205 indicating the crossing of an uplink quality threshold in the uplink direction. Whenever the UE receives an indication of bad quality, it should start the relocation procedure, as described in the next step. Alternatively, UE can use its local measurements to decide to initiate the handover procedure.

Next, the UE decides to initiate an SRNS Relocation from GAN to UTRAN by sending (in Step3) GA-PSR-HANDOVER-INFORMATION message to theGANC6215. Specific criteria for this decision would include the case of the UE leaving GAN coverage (e.g., based on deteriorating WLAN signal quality).

TheGANC6215 selects a target RNC based on the contents of the GA-PSR-HANDOVER-INFORMATION message (e.g., the RNC serving the cell identified by the UE as having the best signal quality). TheGANC6215 sends (in Step4) Relocation Required message to theSGSN6220 containing the selected RNC information.

TheSGSN6220 sends (in Step5) a Relocation Request to thetarget RNC6210. TheRNC6210 performs (in Step6) the necessary allocation of radio and Iu transport resources and returns (in Step7) Relocation Request Acknowledge message to the SGSN. This message contains channelization information needed by UE to access UTRAN.

b) Execution Phase

FIG. 63 illustrates GAN to UTRAN SRNS Relocation Execution Phase in some embodiments. As shown, the followings steps are performed.

TheSGSN6320 begins the Execution Phase by issuing (in Step1) Relocation Command to theGANC6315. The message contains the channel access information in the target UTRAN cell. TheGANC6315 sends (inStep2a) GA-PSR-HANDOVER-COMMAND to theUE6305. This message contains the information from the Relocation Command received inStep1 earlier. The GANC may suspend downlink GTP PDU transfer at this point. If GANC supports Lossless SRNS Relocation and any of existing RABs' QoS requires it, the GANC may initiate (in Step2c) forwarding of GTP PDUs to thetarget RNC6310 via theSGSN6320.

TheGANC6315 also sends (inSteps2band3) Forward SRNS Context to the target RNC via the SGSN. As shown, the GANC sends the Forward SRNS Context message (inStep2b) to the SGSN and the SGSN relays (in Step3) the Forward SRNS Context to the target RNC.

Upon receiving the GA-PSR-HANDOVER-COMMAND, the UE immediately suspends uplink GTP PDU transfer. It immediately begins accessing the UTRAN using indicated channel access parameters in the message. UE's access attempt is detected by the Node B andRNC6310, and is reported (in Step4) to theSGSN6320 via the Relocation Detect message.

The UE completes the lower layer setup and configuration, and sends (in Step5a) the RRC Physical Channel Reconfiguration Complete to thetarget RNC6310. This triggers theRNC6310 to send (inStep5b) the Relocation Complete message toSGSN6320. At this stage, the target RNC assumes the role of SRNC for the UE.

The packet data flow is now (in Step6) active via the UTRAN. Next, the SGSN releases the Iu_PS connection by sending (inStep7a) Iu Release Command message to the GANC, to which GANC responds (inStep7b) with Iu Release Complete message. If the Routing Area of the cell under the target RNC is different from that under the old GANC cell, then theUE6305 performs (in Step8) the Routing Area Update procedure.

U. Short Message Service

GAN provides support for both Circuit Switched and Packet Switched SMS services. GAN-attached UEs will be able to send and receive SMS messages via the GAN.

1. CS-Based SMS

CS-based SMS support in GAN is based on the same mechanism that is utilized for CS mobility management and call control. On the UE side, the SMS layers (including the supporting CM sub layer functions) utilize the services of the MM layer to transfer SMS messages per standard circuit switched UMTS implementation.

The SM-CP protocol is effectively tunneled between the UE and the CN, using GA-CSR UPLINK DIRECT TRANSFER and GA-CSR DOWNLINK DIRECT TRANSFER messages between the UE and the GANC, where the GANC relays the SM-CP messages via RANAP messages for transport over the Iu-cs interface. As with the mobility management and call control procedures, the secure IPSec tunnel and TCP session are used to provide secure and reliable SMS delivery over the IP network.

2. PS-Based SMS

PS-based SMS message transfer is based on the same mechanism as the transfer of the PS mobility management and session management signaling messages. On the UE side, the SMS layers (including the supporting CM sub layer functions) utilize the services of the GA-PSR layer to transfer SMS messages per standard packet switched UMTS implementation. As with mobility management and session management signaling, the secure IPSec tunnel and TCP session is used to provide secure and reliable PS-based SMS delivery over the IP network.

VI. CONFIGURATION INFORMATION

A. GAN UARFCN and Primary Scrambling Code for Handover-to-GAN

In some embodiments, selection of the UMTS Absolute Radio Frequency Channel Number (UARFCN) use the following guidelines:

- 1. The UARFCN should be allocated from the operator's assigned UARFCN values.
- 2. The UARFCN may be desired to be the same unique number across the whole operator network to minimize the RNC configuration effort.
- 3. The Primary Scrambling Code (possible values from 0 to 511) should not be allocated from the operator's in-use values; i.e., codes used by macro cells.
- 4. The Primary Scrambling Code may be desired to be the same unique number across the whole operator network to minimize the RNC configuration effort.

Several options are discussed in more detail below.

1.Option 1

Some embodiments allocate the GAN UARFCN from the DCS band that is being used for GSM. This would result in the DL UARFCNs in the range of 1162 to 1513, inclusive. In this scheme, there are no restrictions in the selection of the specific primary scrambling code (PSC) for GAN—any of the 512 values can be used in the particular UARFCN chosen.

Where initial UMTS deployments are in the 1900 MHz band, an analogous approach may be employed—namely the use of UARFCNs from the 850 MHz band. That would give a GAN UARFCN range of 4357 to 4458, inclusive. Alternatively, UARFCNs from a PCS sub-band doing a non-UMTS technology can also be specified. Again, there are no restrictions in the selection of PSC in a given GAN UARFCN.

2.Option 2

The strategy here is to take advantage of TDD unpaired spectrum and use its UARFCN ranges for GAN purposes. Many operators, as part of the UMTS auction, won a TDD unpaired 5 MHz spectrum, in addition to one or more FDD pairs. The TDD spectrum has remained unused and is likely to remain that way for near foreseeable future.

Even if a given operator does not own any TDD spectrum in a given market, any unused TDD spectrum from any operator in the market can be used since it is a completely harmless interference-free procedure for a UE to do a cell search. Even if a given TDD unpaired 5 MHz is in use in UTRAN-TDD mode, an FDD-only handset is likely fail beyond the initial synchronization at PHY layer. Many handsets planned for near foreseeable future are FDD-only.

If the handsets semantically allow these values, and these UARFCNs are indeed defined in 3GPP, and the infrastructure vendors do allow provisioning of these UARFCN ranges in their systems, then this approach is feasible. The UARFCN ranges in this case are: 9504 to 9596 and 10054 to 10121. As is the case inOption 1, there are no restrictions in PSC selection of GAN.

3.Option 3

This plan calls for use of idle FDD spectrum's UARFCN for GAN purpose. The “idle” spectrum may or may not belong the particular operator. In many parts of Europe and Asia, the FDD spectrums are still unused due to bidders of auction either going out of business or the owners choosing not to deploy services yet due to cost and unavailability of equipment.

VII. IDENTIFIERS IN GAN

A. Identifiers for UEs and Generic IP Access Network

The key UE and generic IP access network addressing parameters are the IMSI associated with the (U)SIM in the terminal, Public IP Address of the UE, and the generic IP access network point of attachment address (AP-ID). The IMSI associated with the (U)SIM is provided by the UE to the GANC during the Registration procedure. The GANC maintains a record for each registered UE. For example, IMSI is used by the GANC to index the appropriate UE record when the GANC receives a RANAP PAGING message.

The Public IP address of UE is the source IP present in the outermost IP header of packets received from the UE by the GANC-SEGW. If available, this identifier may be used by the GANC to support location services and fraud detection or by service providers to signal Managed IP networks IP flows that require special QoS treatment.

The generic IP access network point of attachment address (AP-ID) is provided by the UE to the GANC at Registration. The AP-ID may be used by the GANC to support location services or by the service provider to restrict GAN access to authorized APs.

B. Service Area Identifiers for GAN

1. GAN Service Area for Location Services & Billing

Service Area Identifiers (SAI) in UMTS may be used to perform location-basing routing of a call for services such as: emergency services; operators; announcements and freephone numbers. SAI can be also used by the core network to identify the location of where a call was originated/terminated for charging purposes. The GANC provides a SAI to the core network indicating the Iu-mode GAN service area.

a) Assigning GAN SAI Based on UTRAN/GERAN Location

In the Iu-mode GAN architecture, the UE has a direct IP-based connection to the GANC. The GAN coverage area may overlay the UTRAN/GERAN coverage area. Logical mapping of GAN Cells to a SAI can be completed at various resolutions, for example (but not limited to): (1) a GAN SAI for each UTRAN/GERAN cell, (2) a GAN SAI for each UTRAN/GERAN routing area; or (3) a GAN SAI for each UTRAN/GERAN location area. A single GANC could represent one or more SAI in one or more location areas (LAI).

VIII. ALTERNATIVE EMBODIMENTS

In some embodiments, instead of using separate CSR and PSR protocols, as described in the previous sections, a single protocol, Generic Access Radio Resource Control (GA-RRC) is used. The following sections describe the architecture and messaging features of this protocol layer. Only the features that are different from the previous embodiments are described.

A. Control and User Plane Architecture

The Iu interface standards include support for both ATM and IP-based signaling and user data transport mechanisms.

1. Circuit Switched (CS) Domain

a) CS Domain—Control Plane

FIG. 64 illustrates the GAN architecture in support of the CS Domain control plane in some embodiments. The figure shows different protocol layers for theUE6405,Generic IP Network6410,GANC6415, andMSC6420.FIG. 64 also shows the two interfaces Up6425 and Iu-cs6430. The main features of the GAN CS domain control plane architecture are as follows. Theunderlying Access Layers6435 andTransport IP layer6440 provide the generic connectivity between theUE6405 and theGANC6415. TheIPSec layer6445 provides encryption and data integrity between theUE6405 andGANC6415. TheRemote IP layer6450 is the ‘inner’ IP layer for IPSec tunnel mode and is used by theUE6405 to be addressed by theGANC6415. TheRemote IP layer6450 is configured during the IPSec connection establishment.

In some embodiments asingle TCP connection6455 is used to provide reliable transport for both the GA-RC6460 and GA-RRC6465 signaling between theUE6405 andGANC6415. TheTCP connection6455 is managed by GA-RC6460 and is transported using theRemote IP layer6450.

The Generic Access Resource Control (GA-RC)protocol6460 manages the Up session, including the GAN discovery and registration procedures. The Generic Access Radio Resource Control (GA-RRC)protocol6465 performs functionality equivalent to the UMTS-RRC protocol, using the underlying connection managed by the GA-RC sublayer6460. Note that GA-RRC6465 includes both CS service and PS service-related signaling messages. TheGANC6415 terminates the GA-RRC protocol6465 and inter-works it to theRANAP protocol6470 over the Iu-cs6430 interface. The NAS protocols, such as MM6475 and above, are carried transparently between theUE6405 andMSC6420. In some embodiments, the Iu-cssignaling transport layers6495 are per 3GPP TS 25.412.

b) CS Domain—User Plane

FIG. 65 illustrates the GAN protocol architecture in support of the CS domain user plane in some embodiments. The figure shows different protocol layers for theUE6505,Generic IP Network6510,GANC6515, andMSC6520.FIG. 65 also shows the two interfaces Up6525 and Iu-cs6530. The main features of the GAN CS domain user plane architecture are as follows. Theunderlying Access Layers6535 andTransport IP layer6540 provide the generic connectivity between theUE6505 and theGANC6515.

TheIPSec layer6545 provides encryption and data integrity. CS domain user plane data is transported using the Iu User Plane (Iu UP)protocol6550 running over RTP/UDP (6555 and6560) betweenUE6505 andMSC6520. EachIu UP protocol6550 instance may operate in either transparent or support modes, as described in “UTRAN Iu interface user plane protocols”, 3GPP TS 25.415 standard. The mode choice is indicated to the GANC by the MSC using RANAP and to the UE by the GANC using GA-RRC. Support for the AMR FR codec, as specified in “AMR speech codec; General description”, 3GPP TS 26.071 standard, is mandatory when operating in GAN mode, with support for other codecs being optional. In some embodiments, the Iu-csdata transport layers6595 are per 3GPP TS 25.414.

Some embodiments that utilize GA-RRC protocol implement a protocol stack for the GANC that is different than the protocol stack shown for theGANC6515. In these embodiments, the GANC protocol stack is similar to theGANC1115 protocol stack illustrated inFIG. 11. In these embodiments, the GANC has additional protocol layers Remote IP, UDP, and RTP over theIPSec layer6545. The GANC also has the additional Iu UP protocol layer over the data transport layers6595. Similar to theGANC1115 shown inFIG. 11, the GANC in these embodiments interworks the CS domain user plane between the RTP/UDP and the Iu User Plane protocol.

2. Packet Switched (PS) Domain

a) PS Domain—Control Plane

FIG. 66 illustrates the GAN architecture in support of the PS Domain Control Plane in some embodiments. The figure shows different protocol layers for theUE6605,Generic IP Network6610,GANC6615, andSGSN6620.FIG. 66 also shows the two interfaces Up6625 and Iu-ps6630. The main features of the GAN PS domain control plane architecture are as follows. The functions of GA-RRC6635 and underlying layers are as described in Sub-section VIII.A.1.a: “CS Domain—Control Plane”, above. The GA-RRC protocol6635 performs functionality equivalent to the UTRAN RRC protocol, using the underlying Up session managed by the GA-RC6640. The GA-RRC6635 includes both CS service and PS service-related signaling messages.

TheGANC6615 terminates the GA-RRC protocol6635 and inter-works it to theRANAP protocol6645 over the Iu-ps interface6630. NAS protocols, such as for GMM, SM andSMS6650, are carried transparently between theUE6605 andSGSN6620. In some embodiments, the Iu-pssignaling transport layers6695 are per 3GPP TS 25.412.

b) PS Domain—User Plane

FIG. 67 illustrates the GAN architecture for the PS Domain User Plane in some embodiments. The figure shows different protocol layers for theUE6705,Generic IP Network6710,GANC6715, andSGSN6720.FIG. 67 also shows the two interfaces Up6725 and Iu-ps6730. The main features of the GAN PS domain user plane architecture are as follows. Theunderlying Access Layers6735 andTransport IP6740 layer provides the generic connectivity between theUE6705 and theGANC6715. TheIPSec layer6745 provides encryption and data integrity. The GTP-U6750 protocol operates between theUE6705 and theSGSN6720, transporting the upper layer payload (i.e., PS domain user plane data6755) across theUp6725 and Iu-ps interfaces6730. User data is carried transparently between theUE6705 and core network. In some embodiments, the Iu-ps data transportlower layers6795 are per 3GPP TS 25.414.

Some embodiments that utilize GA-RRC protocol implement a protocol stack for the GANC that is different than the protocol stack shown for theGANC6715. In these embodiments, the GANC protocol stack is similar to theGANC1815 protocol stack illustrated inFIG. 18. In these embodiments, the GANC has additional protocol layers Remote IP, UDP, and GTP-U over theIPSec layer6745. In these embodiments, the GTP-U in the UE and the GTP-U layer over the UDP layer in the GANC is a part of GA-RRC protocol. The GANC also has the additional IP, UDP, and GTP-U layers over the data transportlower layers6795.

3. GA-RC (Generic Access Resource Control)

The GA-RC protocol provides a resource management layer, with the following functions. Discovery and registration with GANC, registration update with GANC, application level keep-alive with GANC, and support for identification of the AP being used for GAN access.

b) States of the GA-RC Sub-Layer

FIG. 68 illustrates the GA-RC sublayer in the UE in some embodiments. As shown, the GA-RC sub-layer in the UE can be in one of two states: GA-RC-DEREGISTERED6805 or GA-RC-REGISTERED6810. In the GA-RC-DEREGISTERED state6805, the UE may be in a GAN coverage area; however, the UE has not registered successfully with the GANC. The UE may initiate the GAN Registration procedure when in the GA-RC-DEREGISTERED state6805. The UE returns to GA-RC-DEREGISTERED state6805 on loss of TCP or IPSec connection or on execution of the GAN De-registration procedure.

In the GA-RC-REGISTEREDstate6810, the UE is registered with the Serving GANC. The UE has an IPSec tunnel and an TCP connection established to the Serving GANC through which the UE may exchange GA-RC or GA-RRC signaling messages with the GANC. While the UE remains in the GA-RC-REGISTEREDstate6805 it performs application level keep-alive with the GANC.

In the GA-RC-REGISTERED state, the UE may be in either UTRAN/GERAN mode6815 orGAN mode6820. The UE (1) may be camped on GERAN or UTRAN and idle, (2) may be active in GERAN or UTRAN (e.g., a GSM RR or a UTRAN RRC connection may be established), (3) may have “roved in” to GAN mode, or (4) may have recently “roved out” of GAN mode (e.g., due to handover from GAN).

4. GA-RRC (Generic Access Radio Resource Control)

The GA-RRC protocol provides a resource management layer, which is a replacement for UTRAN-RRC and provides the following functions: (1) setup of transport channels for CS and PS traffic between the UE and GANC, (2) flow control of PS traffic, (3) CS and PS handover support between UTRAN/GERAN and GAN, (4) direct transfer of NAS messages between the UE and the core network, and (5) other functions such as paging and security configuration.

b) States of the GA-RRC Sub-Layer

The GA-RRC sub-layer in the UE can be in two states, GA-RRC-IDLE6825 or GA-RRC-CONNECTED6830 as illustrated inFIG. 68. The UE enters the GA-RRC-IDLE6825 state when the UE switches the serving RR entity to GA-RRC and the SAP between the NAS and the GA-RRC is activated. This switch may occur only when the GA-RC is in the GA-RC-REGISTERED state. The UE moves from the GA-RRC-IDLE state6825 to the GA-RRC-CONNECTEDstate6830 when the GA-RRC connection is established and returns to GA-RRC-IDLE state when the GA-RRC connection is released. Upon GA-RRC connection release, an indication that no dedicated resources exist is passed to the upper layers. The UE may also enter the GA-RRC-CONNECTED state while in the GA-RC-REGISTERED state in GERAN/UTRAN mode when Handover to GAN is being performed. In the same way, the UE enters the GA-RC-REGISTERED state in GERAN/UTRAN mode from the GA-RRC-CONNECTED state when Handover from GAN is successfully executed.

B. High-Level Procedures

1. GA-RRC Connection Handling

The GA-RRC connection is a logical connection between the UE and the GANC, either for the CS or PS domain. It is established when the upper layers in the UE request GA-RRC to establish a signaling connection and the UE is in idle mode (no RRC connection exists). When a successful response is received from the network, GA-RRC replies to the upper layer that it has entered RRC connected mode. The upper layers have then the possibility to request transmission of NAS messages to the network.

a) GA-RRC Connection Establishment

i) UE Initiated GA-RRC Connection Establishment

FIG. 69 illustrates successful (and unsuccessful) establishment of the GA-RRC Connection when initiated by the UE in some embodiments. TheUE6905 initiates GA-RRC connection establishment by sending (in Step1) the GA-RRC REQUEST message to theGANC6910. This message contains the Establishment Cause indicating the reason for GA-RRC connection establishment. The message also includes the Domain Indicator (CS or PS).GANC6910 signals the successful response to theUE6905 by sending (in Step2) the GA-RRC REQUEST ACCEPT and theUE6905 enters GA-RRC connected mode. Alternatively, theGANC6910 may return (in Step3) a GA-RRC REQUEST REJECT indicating the reject cause.

ii) Network Initiated GA-RRC Connection Establishment

FIG. 70 illustrates successful establishment of the GA-RRC Connection when initiated by the network in some embodiments. TheCN7015 sends (in Step1) a RANAP Paging message to theGANC7010 identified through the last Location Update received by it and includes the TMSI if available. The IMSI of the UE being paged is always included in the request, as is the Domain Indicator (CS or PS). A paging cause may be included.

Next, theGANC7010 identifies the UE registration context using the IMSI provided by theCN7015. It then pages (in Step2) theUE7005 using the GA-RRC PAGING REQUEST message. TheUE7005 responds (in Step3) with a GA-RRC INITIAL DIRECT TRANSFER message containing a NAS message appropriate to the Domain Indicator (CS or PS) and cause. Alternatively, theUE7005 responds (in Step3) with a GA-RRC PAGING RESPONSE message containing a NAS message, the Domain Indicator (i.e., CS or PS) and cause. TheUE7005 enters GA-RRC connected mode. TheGANC7010 establishes an SCCP connection to the CN70015. TheGANC7010 then forwards (in Step4) the NAS message to theCN7015 using the RANAP Initial UE Message. Subsequent NAS messages between the UE and core network will be sent between GANC and CN using the RANAP Direct Transfer message.

b) GA-RRC Connection Release

FIG. 71 shows release of the logical GA-RRC connection between the UE and the GANC in some embodiments. TheCN7115 indicates (in Step1) to theGANC7110 to release the user plane connection allocated to theUE7115, via the RANAP Iu Release Command message. TheGANC7110 confirms (in Step2) resource release toCN7115 using the Tu Release Complete message7125.

Next, theGANC7110 commands (in Step3) theUE7105 to release resources, using the GA-RRC CONNECTION RELEASE message. TheUE7105 confirms (in Step4) resource release to theGANC7110 using the GA-RRC CONNECTION RELEASE COMPLETE message and the GA-RRC state in the UE changes to idle.

3. Security Mode Control

FIG. 72 illustrates the message flow for security mode control in some embodiments. TheCN7215 sends (in Step1) the RANAP Security Mode Command message toGANC7210. This message contains the integrity key (IK) and allowed algorithms, and optionally the encryption key (CK) and allowed algorithms. TheGANC7210 sends (in Step2) the GA-RRC SECURITY MODE COMMAND message to theUE7205. This message indicates the integrity protection and encryption settings (i.e., that are applicable after relocation to UTRAN), and a random number. TheUE7205 stores the information for possible future use after a handover to UTRAN.

Next, theUE7205 computes a MAC based on the random number, the UE IMSI and the integrity key calculated by the UE. TheUE7205 then sends (in Step3) the GA-RRC SECURITY MODE COMPLETE message to signal its selected algorithm and the computed MAC. TheGANC7210 then verifies the MAC using the random number, the UE IMSI and the integrity key provided by theCN7215 instep1. If the GANC verifies the MAC to be correct it sends (in Step4) the Security Mode Complete message to theCN7215. The MAC proves that the identity that is authenticated to the GANC is the same as the identity authenticated to the core network.

4. GA-RRC NAS Signaling Procedures

After GA-RRC connection establishment, NAS signaling may be transferred from CN-to-UE and from UE-to-CN.

a) CN-to-UE NAS Signalling

FIG. 73 illustrates core network to UE NAS signaling of some embodiments. For CN-to-UE NAS signaling, theCore Network7315 sends (in Step1) a NAS PDU to the GANC via the RANAP Direct Transfer message. TheGANC7310 encapsulates (in Step2) the NAS PDU within a GA-RRC DL DIRECT TRANSFER message and forwards the message to theUE7305 via the existing TCP connection.

b) UE-to-CN NAS Signaling

FIG. 74 illustrates the UE to core network NAS signaling of some embodiments. TheUE7405 GA-RRC layer receives a request from the NAS layer to transfer an uplink NAS PDU. Since the MM connection (and hence RR signaling connection) already exists, the UE GA-RRC encapsulates the NAS PDU within a GA-RRC UL DIRECT TRANSFER message and sends (in Step1) the message to theGANC7410. TheGANC7410 relays (in Step2) the received message to theCore Network7415 via the RANAP Direct Transfer message7420.

5. Mobile Originated CS Call

a) UE Terminate Iu UP packet

FIG. 75 illustrates mobile originated CS call procedure in some embodiments. The description of the procedure assumes theUE7505 is in GAN mode; i.e., it has successfully registered with theGANC7510 and GA-RRC is the serving RR entity in theUE7505. It also assumes that no GA-RRC connection exists between theUE7505 and GANC7510 (i.e., GA-RRC-IDLE state). The GA-RRC Connection Establishment procedure is performed (in Step1) as described in Sub-section VIII.B.1.a.i: UE Initiated GA-RRC Connection Establishment, above. Upon request from the upper layers, theUE7505 sends (in Step2) the CM Service Request to theGANC7510 in the GA-RRC INITIAL DIRECT TRANSFER message.

TheGANC7510 establishes an SCCP connection to theCN7515 and forwards (in Step3) the CM Service Request to theCN7515 using the RANAP Initial UE Message. Subsequent NAS messages between theUE7505 andcore network7515 will be sent betweenGANC7510 andCN7515 using the RANAP Direct Transfer message.

TheCN7515 may optionally authenticate (in Step4) theUE7505 using standard UTRAN authentication procedures. TheCN7515 may optionally initiate (in Step5) the Security Mode Control procedure described in Sub-section VIII.B.3: “Security Mode Control”, above.

TheUE7505 sends (in Step6) the Setup message providing details on the call to theCN7515 and its bearer capability and supported codecs. This message is contained within the GA-RRC UL DIRECT TRANSFER between theUE7505 and theGANC7510. TheGANC7510 forwards (in Step6) the Setup message to theCN7515.

TheCN7515 indicates (in Step7) it has received the call setup and it will accept no additional call-establishment information using the Call Proceeding message to theGANC7510. TheGANC7510 forwards (in Step7) this message to theUE7505 in the GA-RRC DL DIRECT TRANSFER message.

TheCN7515 requests (in Step8) theGANC7510 to assign call resources using the RANAP RAB Assignment Request message. TheCN7515 includes the RAB-ID, the CN Transport Layer Address (IP address) and the CN Iu Transport Association (UDP port number) for user data. TheGANC7510 sends (in Step9) the GA-RRC ACTIVATE CHANNEL message to theUE7505 including bearer path setup information received in the RAB Assignment Request message such as: (1) Radio Access Bearer (RAB) parameters; e.g., RAB-ID, UDP port & the IP address for the uplink RTP stream and (2) Iu UP parameters (e.g., Iu UP mode, where support mode is used for AMR voice calls).

Since Iu UP support mode is indicated, theUE7505 sends (in Step10) the Iu UP INITIALISATION packet to the IP address and UDP port indicated in the GA-RRC ACTIVATE CHANNEL message. This message is routed to the core network7515 (e.g., the R4 media gateway). Thecore network7515 responds (in Step11) with the Iu UP INITIALISATION ACK packet. Thecore network7515 sends the message to the source IP address and UDP port number of the received INITIALISATION packet.

TheUE7505 sends (in Step12) the GA-RRC ACTIVATE CHANNEL ACK to theGANC7510. The GANC signals (in Step13) to theCN7515 that the RAB has been established by sending a RANAP RAB Assignment Response message. TheGANC7510 signals (in Step14) the completion of the RAB establishment to theUE7505 with the GA-RRC ACTIVATE CHANNEL COMPLETE message.

An end-to-end audio path now exists between theUE7505 and theCN7515. TheUE7505 can now connect the user to the audio path. TheCN7515 signals to theUE7505, with the Alerting message, that the called party is ringing. The message is transferred (in Step15) to theGANC7510 and GANC forwards (in Step15) the message to theUE7505 in the GA-RRC DL DIRECT TRANSFER.

When theUE7505 has not connected the audio path to the user, it generates ring back to the calling party. Otherwise, the network-generated ring back will be returned to the calling party. TheCN7515 signals that the called party has answered, via the Connect message. The message is transferred (in Step16) to theGANC7510 and GANC forwards (in Step16) the message to theUE7505 in the GA-RRC DL DIRECT TRANSFER7595. TheUE7505 connects the user to the audio path. If theUE7505 is generating ring back, it stops and connects the user to the audio path.

TheUE7505 sends (in Step17) the Connect Ack message in response, and the two parties are connected for the voice call. This message is contained within the GA-RRC UL DIRECT TRANSFER between theUE7505 and theGANC7510. The GANC forwards (in Step17) the Connect Ack message to theCN7515. Bi-directional voice traffic flows (in Step18) between theUE7505 andCN7515 through theGANC7510.

b) GANC Terminates Iu UP packet

Some embodiments utilize an alternative procedure for the mobile originated CS call using RRC protocol.FIG. 76 illustrates steps performed during a mobile originated CS call in these embodiments. The procedure assumes that the UE is in GAN mode; i.e., it has successfully registered with the GANC and GA-RRC is the serving RR entity for CS services in the UE. It also assumes that no GA-RRC signaling connection exists between the UE and GANC (i.e., GA-RRC-IDLE state). As shown, the GA-RRC Connection Establishment procedure is performed (in Step1). In some embodiments, this procedure is performed. Next, theUE7605 sends the CM Service Request message to theGANC7610 within the GA-RRC UL DIRECT TRANSFER message.

Next, theGANC7610 establishes an SCCP connection to thecore network CN7615 and forwards (in Step3) the NAS PDU (i.e., the CM Service Request message) to thecore network CN7615 using the RANAP Initial UE Message. The message includes the Domain Indicator set to value ‘CS domain’. Subsequent NAS messages between the UE and core network CN will be sent between GANC and core network CN using the RANAP Direct Transfer message.

Thecore network CN7615 may optionally authenticate (in Step4) the UE using standard UTRAN authentication procedures. Thecore network CN7615 may optionally initiate (in Step5) the Security Mode Control procedure. TheUE7605 sends (in Step6) the Setup message providing details on the call to the core network CN and its bearer capability and supported codecs. This message is contained within the GA-RRC UL DIRECT TRANSFER between the UE and the GANC. The GANC forwards the Setup message to the core network CN.

Next, thecore network CN7615 indicates (in Step7) it has received the call setup and it will accept no additional call-establishment information using the Call Proceeding message to the GANC. The GANC forwards (in Step7) this message to the UE in the GA-RRC DL DIRECT TRANSFER message.

Thecore network CN7615 requests (in Step8) theGANC7610 to assign call resources using the RANAP RAB Assignment Request message. Thecore network CN7615 includes the RAB-ID, the CN Transport Layer Address and the CN Iu Transport Association for user data, and an indication that Iu UP support mode is required, among other parameters.

TheGANC7610 then sends (in Step9) the GA-RRC ACTIVATE CHANNEL message to theUE7605 including bearer path setup information such as: (1) Channel mode, (2)

Multi-rate codec configuration, (3) UDP port & the IP address for the uplink RTP stream, and (4)

Voice sample size.

Next, theUE7605 sends (in Step10) the GA-RRC ACTIVATE CHANNEL ACK to theGANC7610 indicating the UDP port for the downlink RTP stream. Since Iu UP support mode is indicated by the core network CN instep8, theGANC7610 sends (in Step11) the Iu UP INITIALIZATION packet to the core network CN.

In response, the core network CN responds (in Step12) with the Iu UP INITIALISATION ACK packet. TheGANC7610 signals (in Step13) the completion of the RAB establishment to theUE7605 with the GA-RRC ACTIVATE CHANNEL COMPLETE message. Alternatively, Steps11 and12 may occur beforeStep9.

TheGANC7610 signals to thecore network CN7615 that the RAB has been established by sending (in Step14) a RANAP RAB Assignment Response message. Thecore network CN7615 signals to theUE3505, with the Alerting message, that the called party is ringing. The message is transferred (in Step15) to theGANC7610 and GANC forwards (in Step15) the message to theUE7605 in the GA-RRC DL DIRECT TRANSFER. When the UE has not connected the audio path to the user, it generates ring back to the calling party. Otherwise, the network-generated ring back will be returned to the calling party.

Next, thecore network CN7615 signals that the called party has answered, via the Connect message. The message is transferred (in Step16) to theGANC7610 and GANC forwards (in Step16) the message to the UE in the GA-RRC DL DIRECT TRANSFER. The UE connects the user to the audio path. If the UE is generating ring back, it stops and connects the user to the audio path.

TheUE7605 then sends (in Step17) the Connect Ack message in response, and the two parties are connected for the voice call. This message is contained within the GA-RRC UL DIRECT TRANSFER between the UE and the GANC. The GANC forwards the Connect Ack message to the core network CN. At this time, bi-directional voice traffic flows (in Step18) between theUE7605 andcore network CN7615 through theGANC7610.

6. Mobile Terminated CS Call

FIG. 77 illustrates mobile terminated CS call procedure in some embodiments. The description of the procedure assumes theUE7705 is in GAN mode; i.e., it has successfully registered with theGANC7710 and GA-RRC is the serving RR entity in theUE7705. It also assumes that no GA-RRC connection exists between theUE7705 and GANC7710 (i.e., GA-RRC-IDLE state).

A mobile-terminated call arrives at theCN7715. TheCN7715 sends (in Step1) a RANAP Paging message to theGANC7710 identified through the last Location Update received by it and includes the TMSI if available. The IMSI of the mobile being paged is always included in the request. TheGANC7710 identifies the UE registration context using the IMSI provided by theCN7715. The GANC then pages (in Step2) theUE7705 using the GA-RRC PAGING REQUEST message. The message includes the TMSI, when available in the request from theCN7715. Otherwise, the message includes only the IMSI of theUE7705.

TheUE7705 responds (in Step3) with a GA-RRC INITIAL DIRECT TRANSFER message containing the Paging Response. TheUE7705 enters GA-RRC connected mode. TheGANC7710 establishes an SCCP connection to theCN7715. TheGANC7710 then forwards (in Step4) the paging response to theCN7715 using the RANAP Initial UE Message. Subsequent NAS messages between theUE7705 andcore network7715 will be sent betweenGANC7710 andCN7715 using the RANAP Direct Transfer message.

TheCN7715 may optionally authenticate (in Step5) theUE7705 using standard UTRAN authentication procedures. TheCN7715 may optionally update (in Step6) the security configuration in theUE7705, via theGANC7710, as described in Sub-section VIII.B.3: “Security Mode Control”, above. TheCN7715 initiates call setup using the Setup message sent (in Step7) to theUE7705 viaGANC7710. GANC forwards (in Step7) this message to theUE7705 in the GA-RRC DL DIRECT TRANSFER message.

TheUE7705 responds (in Step8) with Call Confirmed using the GA-RRC UL DIRECT TRANSFER after checking it's compatibility with the bearer service requested in the Setup and modifying the bearer service as needed. If the Setup included the signal information element, theUE7705 alerts the user using the indicated signal, else theUE7705 alerts the user after the successful configuration of the user plane. TheGANC7710 forwards (in Step8) the Call Confirmed message to theCN7715. TheCN7715 initiates (in Step9) the assignment procedure with theGANC7710, which triggers the setup of the RTP stream (voice bearer channel) between theGANC7710 andUE7705.

TheUE7705 signals (in Step10) that it is alerting the user, via the Alerting message contained in the GA-RRC UL DIRECT TRANSFER. TheGANC7710 forwards (in Step10) the Alerting message to theCN7715. TheCN7715 sends a corresponding alerting message to the calling party. TheUE7705 signals (in Step11) that the called party has answered, via the Connect message contained in the GA-RRC UL DIRECT TRANSFER. TheGANC7710 forwards (in Step11) the Connect message to theCN7715. TheCN7715 sends a corresponding Connect message to the calling party and through connects the audio. TheUE7705 connects the user to the audio path.

TheCN7715 acknowledges (in Step12) via the Connect Ack message to theGANC7710.GANC7710 forwards (in Step12) this message to theUE7705 in the GA-RRC DL DIRECT TRANSFER. The two parties on the call are connected on the audio path. Bi-directional voice traffic flows (in Step13) between theUE7705 andCN7715 through theGANC7710.

7. CS Call Clearing

FIG. 78 illustrates call clearing initiated by the UE in some embodiments. As shown, theUE7805 sends (in Step1) the Disconnect message to theCN7815 to release the call. This message is contained in the GA-RRC UL DIRECT TRANSFER message betweenUE7805 andGANC7810. TheGANC7810 forwards (in Step1) the Disconnect message to the CN7815 (i.e., using the RANAP Direct Transfer message).

TheCN7815 responds (in Step2) with a Release message to theGANC7810. TheGANC7810 forwards (in Step2) this message to theUE7805 using the GA-RRC DL DIRECT TRANSFER message.

TheUE7805 responds (in Step3) with the Release Complete message. This message is contained within the GA-RRC UL DIRECT TRANSFER message betweenUE7805 andGANC7810. TheGANC7810 forwards (in Step3) the Disconnect message to theCN7815. TheCN7815 triggers (in Step4) the release of connection as described in Subsection VIII.B.1.b: “GA-CSR Connection Release”.

8. CS Handover

a) CS handover from GERAN to GAN

i) UE terminates Iu UP Packet

FIG. 79 illustrates the CS Handover from GERAN to GAN procedure in some embodiments. The description of the GERAN to GAN handover procedure assumes the following: (1) the UE is on an active call on the GERAN; (2) the UE mode selection is GAN-preferred, or if GERAN/UTRAN-preferred, the RxLev from the current serving cell drops below a defined threshold, in some embodiments this threshold can be specified as a fixed value, or provided by the GERAN BSS to the UE in dedicated mode; (3) the UE has successfully registered with a GANC, allowing the UE to obtain GAN system information; and (4) the GERAN provides information on neighboring 3G cells such that one of the cells in the 3G neighbor list matches the 3G cell information associated with the GANC, as provided in the AS-related component of the system information obtained from the GANC.

The UE begins to include (in Step1) GAN cell information in the Measurement Report message to the GERAN. The UE reports the highest signal level for the GAN cell. This is not the actual measured signal level on GAN, rather an artificial value (i.e., RxLev=63), allowing the UE to indicate preference for the GAN.

Based on UE measurement reports and other internal algorithms, the GERAN BSC decides to handover to the GAN cell. TheBSC7920 starts the handover preparation by sending (in Step2) a Handover Required message to theCN7915, identifying thetarget 3G RNC (GANC)7910. TheCN7915 requests (in Step3) thetarget GANC7910 to allocate resources for the handover using the Relocation Request message. TheUE7905 is identified by the included IMSI parameter.

TheGANC7910 sends (in Step4) the GA-RRC ACTIVATE CHANNEL message to theUE7905 including bearer path setup information received in the Relocation Request message, such as: (1) UDP port & the IP address for the uplink RTP stream, (2) Radio Access Bearer (RAB) parameters, and (3) Iu UP parameters (e.g., Iu UP mode, where support mode is used for AMR voice calls).

Since Iu UP support mode is indicated, theUE7905 sends (in Step5) the Iu UP INITIALISATION packet to the IP address and UDP port indicated in the GA-RRC ACTIVATE CHANNEL message. This message is routed to the core network7915 (e.g., the R4 media gateway).

Thecore network7915 responds (in Step6) with the Iu UP INITIALISATION ACK packet. Thecore network7915 sends the message to the source IP address and UDP port number of the received INITIALISATION packet. TheUE7905 sends (in Step7) the GA-RRC ACTIVATE CHANNEL ACK to theGANC7910. TheGANC7910 builds a Handover to UTRAN Command message and sends (in Step8) it to theCN7915 through the Relocation Request Acknowledge message.

TheGANC7910 signals (in Step9) the completion of the RAB establishment to theUE7905 with the GA-RRC ACTIVATE CHANNEL COMPLETE message. An end-to-end audio path now exists between theUE7905 and theCN7915. TheCN7915 forwards (in Step10) the Handover to UTRAN Command message to theGERAN BSC7920 in the BSSMAP Handover Command message, completing the handover preparation.

TheGERAN BSC7920 sends (in Step11) the Intersystem to UTRAN Handover Command message, containing the Handover to UTRAN Command message, to the UE to initiate handover to GAN. The UE does not switch its audio path from GERAN to GAN until handover completion (i.e., until it sends the GA-RRC HANDOVER COMPLETE message) to keep the audio interruption short.

The UE accesses theGANC7910 using (in Step12) the GA-RRC HANDOVER ACCESS message, and provides the entire Intersystem to UTRAN Handover Command message received from GERAN. TheGANC7910 indicates (in Step13) to theCN7915 that it has detected the UE, using Relocation Detect message. TheCN7915 can optionally now switch the user plane from the source GERAN to the target GAN. Bi-directional voice traffic is now flowing (in Step14) between the UE andCN7915, viaGANC7910.

The UE transmits (in Step15) the GA-RRC HANDOVER COMPLETE message to indicate the completion of the handover procedure at its end. It switches the user from the GERAN user plane to the GAN user plane.

Thetarget GANC7910 indicates (in Step16) the handover is complete, using the Relocation Complete message. If it had not done so before, theCN7915 now switches the user plane from source GERAN to target GAN.

Finally, theCN7915 tears (in Step17) down the connection to the source GERAN, using Clear Command message. The source GERAN confirms (in Step18) the release of GERAN resources allocated for this call, using Clear Complete message.

ii) GANC Terminates Iu UP Packet

FIG. 80 illustrates an alternative procedure for CS handover from GERAN to GAN in some embodiments. The description of the GERAN to GAN handover procedure assumes the following: (1) the UE is on an active call on the GERAN, (2) the UE mode selection is GAN-preferred, or if GERAN/UTRAN-preferred, the RxLev from the current serving cell drops below a defined threshold. In some embodiments, this threshold can be specified as a fixed value, or provided by the GERAN BSS to the UE in dedicated mode, (3) the UE has successfully registered with a GANC, allowing the UE to obtain GAN system information, and (4) the GERAN provides information on neighboring 3G cells such that one of the cells in the 3G neighbor list matches the 3G cell information associated with the GANC, as provided in the AS-related component of the system information obtained from the GANC. As shown, theUE8005 begins to include GAN cell information in the Measurement Report message to theGERAN BSC8015. TheUE8005 reports the highest signal level for the GAN cell. This is not the actual measured signal level on GAN, rather an artificial value (e.g., RxLev=63), allowing the UE to indicate preference for the GAN.

Based on UE measurement reports and other internal algorithms, theGERAN BSC8015 decides to handover to the GAN cell. TheBSC8015 starts the handover preparation by sending (in Step2) a Handover Required message to the core network CN (8020), identifying thetarget 3G RNC (GANC).

The core network CN (8020) requests (in Step3) thetarget GANC8010 to allocate resources for the handover using the Relocation Request message. The UE is identified by the included IMSI parameter.

Since Iu UP support mode is indicated, theGANC8010 sends (in Step4) the Iu UP INITIALISATION packet to the core network CN. The core network CN responds (in Step5) with the Iu UP INITIALISATION ACK packet.

TheGANC8010 builds a Handover to UTRAN Command message and sends it (in Step6) to thecore network CN8020 through the Relocation Request Acknowledge message. The core network CN forwards (in Step7) the Handover to UTRAN Command message to theGERAN BSC8015 in the BSSMAP Handover Command message, completing the handover preparation.

Next, theGERAN BSC8015 sends (in Step8) the Intersystem to UTRAN Handover Command message, containing the Handover to UTRAN Command message, to theUE8005 to initiate handover to GAN. The UE does not switch its audio path from GERAN to GAN until handover completion (i.e., until it sends the GA-RRC HANDOVER COMPLETE message) to keep the audio interruption short.

TheUE8005 accesses (in Step9) theGANC8010 using the GA-RRC HANDOVER ACCESS message, and provides the entire Intersystem to UTRAN Handover Command message received from GERAN. TheGANC8010 sends (in Step10) the GA-RRC ACTIVATE CHANNEL message to theUE8005 including bearer path setup information such as: (1) Channel mode, (2) Multi-rate codec configuration, (3) UDP port & the IP address for the uplink RTP stream, and (4) Voice sample size.

Next, theUE8005 sends (in Step11) the GA-RRC ACTIVATE CHANNEL ACK to theGANC8010 indicating the UDP port for the downlink RTP stream. TheGANC8010 signals (in Step11) the completion of the RAB establishment to theUE8005 with the GA-RRC ACTIVATE CHANNEL COMPLETE message.

TheUE8005 transmits (in Step13) the GA-RRC HANDOVER COMPLETE message to indicate the completion of the handover procedure at its end. It switches the user from the GERAN user plane to the GAN user plane. TheGANC8010 indicates (in Step14) to the core network CN (8020) that it has detected the UE, using Relocation Detect message. The CN can optionally now switch the user plane from the source GERAN to the target GAN.

Bi-directional voice traffic is now (in Step15) flowing between theUE8005 andcore network CN8020, viaGANC8010. Thetarget GANC8010 indicates (in Step16) the handover is complete, using the Relocation Complete message. If it had not done so before, the CN now switches the user plane from source GERAN to target GAN.

The CN tears down (in Step17) the connection to the source GERAN, using Clear Command message. Finally, thesource GERAN8015 confirms (in Step18) the release of GERAN resources allocated for this call, using Clear Complete message.

b) CS handover from UTRAN to GAN

i) UE Terminate Iu UP Packet

The description of the UTRAN to GAN Handover procedure assumes the following: (1) the UE is on an active call on the UTRAN; (2) the UE has been ordered by the RNC to make inter-frequency measurements. When the UE is in GAN preferred mode with an Event2A configured, the UE handles parameters associated with the Event2A in a GAN specific manner (as described in 3GPP TS 25.331) for the reporting of the GAN. When the UE is in GERAN/UTRAN preferred mode and an Event2A has been configured for the GAN cell, the UE shall only send a measurement about the GAN cell, when this event is triggered and no UTRAN cells from the neighbour cell list of the UE satisfy the triggering condition of this Event (as described in 3GPP TS 25.331); and (3) the UTRAN provides information on neighbouring cells such that one of the cells in the neighbour list matches the cell associated with the GANC, as provided in the AS-related component of the system information obtained from GANC.

FIG. 81 illustrates the CS Handover from UTRAN to GAN procedure in some embodiments. The UE begins to include (in Step1) information about a GAN cell in the Measurement Report message sent to theRNC8120. The UE reports the highest signal level for the GAN cell. This is not the actual measured signal level on the GAN, rather an artificial value allowing the UE to indicate preference for the GAN.

Based on UE measurement reports and other internal algorithms, theRNC8120 decides to initiate handover to the GAN cell. TheRNC8120 starts the preparation phase of the Relocation procedure by sending (in Step2) a Relocation Required message to theCN8115, identifying the target (EGAN) cell.

TheCN8115 requests (in Step3) thetarget GANC8110 to allocate resources for the handover using the Relocation Request message. TheUE8105 is identified by the included IMSI parameter.

TheGANC8110 sends (in Step4) the GA-RRC ACTIVATE CHANNEL message to theUE8105 including bearer path setup information received in the Relocation Request message, such as: (1) UDP port & the IP address for the uplink RTP stream, (2) Radio Access Bearer (RAB) parameters, and (3) Iu UP parameters (e.g., Iu UP mode, where support mode is used for AMR voice calls).

Since Iu UP support mode is indicated, theUE8105 sends (in Step5) the Iu UP INITIALISATION packet to the IP address and UDP port indicated in the GA-RRC ACTIVATE CHANNEL message. This message is routed to the core network8115 (e.g., the R4 media gateway).

Thecore network8115 responds (in Step6) with the Iu UP INITIALISATION ACK packet. Thecore network8115 sends the message to the source IP address and UDP port number of the received INITIALISATION packet. TheUE8105 sends (in Step7) the GA-RRC ACTIVATE CHANNEL ACK to theGANC8110.

Thetarget GANC8110 acknowledges (in Step8) the handover request message, using Relocation Request Acknowledge message, indicating it can support the requested handover, and including a Physical Channel Reconfiguration message that indicates the radio channel to which theUE8105 should be directed.

TheGANC8110 signals (in Step9) the completion of the RAB establishment to theUE8105 with the GA-RRC ACTIVATE CHANNEL COMPLETE message. An end-to-end audio path now exists between theUE8105 and theCN8115. TheCN8115 sends (in Step10) the Relocation Command message to theRNC8120, completing the relocation preparation.

TheRNC8120 sends (in Step11) the PHYSICAL CHANNEL RECONFIGURATION message to the UE to initiate handover to GAN. The UE does not switch its audio path from UTRAN to GAN until handover completion (i.e., until it sends the GA-RRC HANDOVER COMPLETE message) to keep the audio interruption short. The UE accesses (in Step12) theGANC8110 using the GA-RRC HANDOVER ACCESS message, and provides the entire PHYSICAL CHANNEL RECONFIGURATION message received fromRNC8120.

TheGANC8110 indicates (in Step13) to theCN8115 that it has detected the UE, using Relocation Detect message. TheCN8115 can optionally now switch the user plane from thesource RNC8120 to thetarget GANC8110. Bi-directional voice traffic is now flowing (in Step14) between the UE andCN8115, viaGANC8110.

The UE transmits (in Step15) the GA-RRC HANDOVER COMPLETE to indicate the completion of the handover procedure from its perspective. It switches the user from the UTRAN user plane to the GAN user plane. Thetarget GANC8110 indicates (in Step16) the handover is complete, using the Relocation Complete message. If it has not done so before, theCN8115 now switches the user plane fromsource RNC8120 to targetGANC8110.

Finally, theCN8115 tears (in Step17) down the connection to thesource RNC8120, using Iu Release Command. Thesource RNC8120 confirms (in Step18) the release of UTRAN resources allocated for this call, using Iu Release Complete.

ii) GANC terminates Iu UP Packet

FIG. 82 illustrates an alternative procedure for CS handover from UTRAN to GAN using RRC protocol in some embodiments. The description of the UTRAN to GAN Handover procedure assumes the following: (1) the UE is on an active call on the UTRAN, (2) the UE has been ordered by the RNC to make inter-frequency measurements (i.e., if the GAN cell has been allocated a different frequency value than is used in the UTRAN), (a) if the UE is in GAN preferred mode with an Event2A configured, the UE handles parameters associated with the Event2A in a GAN specific manner for the reporting of the EGAN, (b) when the UE is in GERAN/UTRAN preferred mode and an event2A has been configured for the GAN cell, the UE shall only send a measurement about the GAN cell, when this event is triggered and no UTRAN cells from the neighbor cell list of the UE satisfy the triggering condition of this Event (as described in 3GPP TS 25.331), (3) the UTRAN provides information on neighboring cells such that one of the cells in the neighbor list matches the cell associated with the GANC, as provided in the AS-related component of the system information obtained from GANC.

As shown inFIG. 82, theUE8205 begins to include information about a GAN cell in the Measurement Report message sent (in Step1) to theRNC8215. TheUE8205 reports the highest signal level for the GAN cell. This is not the actual measured signal level on the GAN, rather an artificial value allowing theUE8205 to indicate preference for the GAN.

Based on UE measurement reports and other internal algorithms, theRNC8215 decides to initiate handover to the GAN cell. TheRNC8215 starts the preparation phase of the Relocation procedure by sending (in Step2) a Relocation Required message to the core network CN, identifying the target (GAN) cell.

Next, steps3 to5 shown inFIG. 82 are performed similar to steps3-5 for CSR GERAN to GAN Handover “GANC Terminates Iu UP Packets” Sub-section described above, except that the messages are RRC messages (instead of CSR). Thetarget GANC8210 acknowledges (in Step6) the handover request message, using Relocation Request Acknowledge message, indicating it can support the requested handover, and including a Physical Channel Reconfiguration message that indicates the radio channel to which the UE should be directed.

Next, thecore network CN8220 sends (in Step7) the Relocation Command message to theRNC8215, completing the relocation preparation. TheRNC8215 sends (in Step8) the PHYSICAL CHANNEL RECONFIGURATION message to theUE8205 to initiate handover to GAN. The UE does not switch its audio path from UTRAN to GAN until handover completion (i.e., until it sends the GA-RRC HANDOVER COMPLETE message) to keep the audio interruption short.

Next, Steps9-16 shown inFIG. 82 are performed similar to Steps9-16 for CSR GERAN to GAN Handover in “GANC Terminates Iu UP Packets” Sub-section described above, except that Steps9-16 inFIG. 82 utilize RRC protocol instead of CSR protocol. Next, thecore network CN8220 tears down (in Step17) the connection to the source RNC, using Iu Release Command. Finally, thesource RNC8215 confirms (in Step18) the release of UTRAN resources allocated for this call, using Iu Release Complete.

c) CS Handover from GAN to GERAN

The procedure description in this sub-clause assumes the following: (1) the UE is on an active call on the EGAN; and (2) the GERAN becomes available and (i) the UE mode selection is GERAN/UTRAN-preferred, or (ii) the UE mode selection is GAN-preferred and the UE begins to leave GAN coverage, based on its local measurements, received RTCP reports, as well as any uplink quality indications received from the GANC.

The handover from GAN to GERAN procedure is always triggered by the UE.

FIG. 83 illustrates the CS handover from GAN to GERAN procedure in some embodiments. TheGANC8310 may send (in Step1) a GA-RRC UPLINK QUALITY INDICATION if there is a problem with the uplink quality for the ongoing call. Uplink Quality Indication is information sent by theGANC8310 to theUE8305 indicating the crossing of an uplink quality threshold in the uplink direction. Whenever theUE8305 receives an indication of bad quality, it should start the handover procedure, as described in the next step. Alternatively,UE8305 can use its local measurements or received RTCP reports, to decide to initiate the handover procedure.

TheUE8305 sends (in Step2) the GA-RRC HANDOVER INFORMATION message to theGANC8310 indicating the Channel Mode and a list of target GERAN cells, identified by CGI, in order of preference (e.g. ranked by C1 path loss parameter) for handover, and includes the received signal strength for each identified GERAN cell. This list is the most recent information available from the GSM RR subsystem. In addition, the GA-RRC HANDOVER INFORMATION message may include a list of target UTRAN cells ranked in order of preference for handover, and the received signal strength for each identified UTRAN cell.

If theServing GANC8310 selects a target GERAN cell, the handover to GERAN procedure is performed. TheServing GANC8310 starts the handover preparation by signaling (in Step3) to theCN8315 the need for handover, using Relocation Required, and including the GERAN cell list provided by theUE8305. TheGANC8310 may include only a subset of the cell list provided by theUE8305.

TheCN8315 selects a target GERAN cell and requests (in Step4) it to allocate the necessary resources, using Handover Request. The target GERAN builds a Handover Command message providing information on the channel allocated and sends (in Step5) it to theCN8315 through the Handover Request Acknowledge message.

TheCN8315 signals (in Step6) theGANC8310 to handover theUE8305 to the GERAN, using Relocation Command message, ending the handover preparation phase.GANC8310 transmits (in Step7) the GA-RRC HANDOVER COMMAND to theUE8305 including the details sent by the GERAN on the target resource allocation. TheUE8305 transmits (in Step8) the Um: Handover Access containing the handover reference element to allow the target GERAN to correlate this handover access with the Handover Command message transmitted earlier to theCN8315 in response to the Handover Required.

The target GERAN confirms (in Step9) the detection of the handover to theCN8315, using the Handover Detect message. TheCN8315 may at this point switch (in Step10) the user plane to the target BSS. The GERAN provides (in Step11) Physical Information to the UE8305 (i.e., Timing Advance) to allow theUE8305 to synchronize with the GERAN. TheUE8305 signals (in Step12) to the GERAN that the handover is completed, using Handover Complete.

The GERAN confirms (in Step13) to theCN8315 the completion of the handover, via Handover Complete message. TheCN8315 may use the target CGI used in the Handover procedure for charging purposes. Bi-directional voice traffic is now flowing (in Step14) between theUE8305 andCN8315, via the GERAN.

On receiving the confirmation of the completion of the handover, theCN8315 indicates (in Step15) to theGANC8310 to release any resources allocated to theUE8305, via the Iu Release Command.GANC8310 commands (in Step16) theUE8305 to release resources, using the GA-RRC RELEASE message.GANC8310 confirms (in Step17) resource release toCN8315 using the Iu Release Complete message.

TheUE8305 confirms (in Step18) resource release to theGANC8310 using the GA-RRC RELEASE COMPLETE message. TheUE8305 may finally deregister (in Step19) from theGANC8310, using GA-RC DEREGISTER message.

d) CS Handover from GAN to UTRAN

The procedure description in this sub-clause assumes the following: (1) the UE is on an active call on the GAN; (2) the UE is capable of operating in all of the GAN, GERAN and UTRAN modes; and (3) the UTRAN becomes available and (i) the UE is in GERAN/UTRAN-preferred mode, or (ii) the UE mode selection is GAN preferred and begins to leave GAN coverage, based on its local measurements, received RTCP reports, as well as any uplink quality indications received from the GANC.

FIG. 84 illustrates the CS handover from GAN to UTRAN procedure in some embodiments. The handover from GAN procedure is always triggered by theUE8405. TheGANC8410 may send (in Step1) a GA-RRC UPLINK QUALITY INDICATION if there is a problem with the uplink quality for the ongoing call. Uplink Quality Indication is information sent by theGANC8410 to theUE8405 indicating the crossing of an uplink quality threshold in the uplink direction. Whenever theUE8405 receives an indication of bad quality, it should start the handover procedure, as described in the next step. Alternatively,UE8405 can use its local measurements or received RTCP reports, to decide to initiate the handover procedure.

TheUE8405 sends (in Step2) the GA-RRC HANDOVER INFORMATION message to theServing GANC8410 indicating the Channel Mode and a list of candidate target UTRAN and GERAN cells, in order of preference for handover, and includes the received signal strength for each identified cell. The UTRAN cells are identified by the PLMN ID, the LAC and the 3G Cell identity (defined in 3GPP TS 25.331).

If theServing GANC8410 selects UTRAN as the target RAT, the handover to UTRAN procedure is performed. TheServing GANC8410 starts the handover preparation by signaling (in Step3) to theCN8415 the need for handover, using Relocation Required and including the UTRAN cell list provided by theUE8405. TheGANC8410 may include only a subset of the cell list provided by theUE8405.

TheCN8415 starts the handover procedure towards thetarget RNC8420 identified by theServing GANC8410. TheCN8415 requests (in Step4) from thetarget RNC8420 to allocate the necessary resources using Relocation Request. Thetarget RNC8420 builds a Physical Channel Reconfiguration message providing information on the allocated UTRAN resources and sends (in Step5) it to theCN8415 through the Relocation Request Acknowledge message.

TheCN8415 signals (in Step6) theServing GANC8410 to handover theUE8405 to the UTRAN, using Relocation Command message (which includes the Physical Channel Reconfiguration message), ending the handover preparation phase. TheServing GANC8410 transmits (in Step7) the GA-RRC HANDOVER COMMAND to theUE8405 including the details sent by the UTRAN on the target resource allocation.

Target RNS achieves (in Step8) uplink synchronization on the Uu interface. Thetarget RNC8420 confirms (in Step9) the detection of the handover to theCN8415, using the Relocation Detect message. TheCN8415 may at this point switch (in Step10) the user plane to the target RNS. TheUE8405 signals (in Step11) to the UTRAN that the handover is completed, using Handover to UTRAN Complete.

The UTRAN confirms (in Step12) to theCN8415 the completion of the handover, via Relocation Complete message. If the user plane has not been switched inStep10, theCN8415 switches the user plane to the target RNS. Bi-directional voice traffic is now flowing (in Step13) between theUE8405 andCN8415, via the UTRAN.

On receiving the confirmation of the completion of the handover, theCN8415 indicates (in Step14) to theServing GANC8410 to release any resources allocated to theUE8405, via the Iu Release Command. TheServing GANC8410 commands (in Step15) theUE8405 to release resources, using the GA-RRC RELEASE message.

TheServing GANC8410 confirms (in Step16) resource release toCN8415 using the Iu Release Complete message. TheUE8405 confirms (in Step17) resource release to theServing GANC8410 using the GA-RRC RELEASE COMPLETE message. TheUE8405 may finally deregister (in Step18) from theServing GANC8410, using GA-RC DEREGISTER message.

9. GA-RRC Packet Transport Channel Management Procedures

The GA-RRC Packet Transport Channel (GA-RRC PTC) provides the association between the UE and the network for the transport of GPRS user data over the Up interface (i.e., via the GAN in Iu-mode). The PTC uses the GTP-U protocol running over UDP transport. The endpoint addresses of the PTC are identified by the IP addresses and UDP ports assigned to the PTC in the UE and network during the PTC activation procedure. The UDP port number for GTP-U is as defined in 3GPP TS 25.414. Multiple PTC instances between a UE and the network may be activated at the same time, using the same endpoint addresses. Each PTC instance is assigned unique GTP-U Tunnel Endpoint IDs (one on the UE and one on the network) during the activation procedure. The UE and GANC manage the activation and deactivation of the PTC instances based on the requests for data transfer and the configurable PTC Timer.

a) States of the GA-RRC Packet Transport Channel

The UE in the GA-RRC-CONNECTED state can be in one of two PTC substates: PTC-STANDBY or PTC-ACTIVE. PTC-STANDBY: this is the initial/default PTC substate of the UE when in the GA-RRC-CONNECTED state in GAN mode. The UE is not able to send or receive GPRS user data to or from the network. The UE needs to activate the PTC before sending any GPRS user data. When the UE successfully establishes a PTC, the UE transitions to the PTC-ACTIVE substate. PTC-ACTIVE the UE is in the GA-RRC-CONNECTED state and the PTC is active between the UE and the network and the UE is able to send and receive GPRS user data to and from the network. The following are the possible triggers for GA-RRC PTC activation on the UE side: (1) The UE initiates the uplink user data transfer, and (2) the GANC initiates PTC activation; i.e., the UE receives a GA-RRC-ACTIVATE-PTC-REQUEST message from the GANC.

On successful PTC activation and in parallel with transition to the PTC-ACTIVE substate, the UE starts the PTC Timer. When the PTC Timer expires, the UE sends a message to the GANC to initiate PTC deactivation. On successful PTC deactivation, the UE transitions to PTC-STANDBY substate. At any time while in the GA-RRC-CONNECTED state and the PTC-ACTIVE substate, the UE may receive the GA-RRC RELEASE message. In addition to requesting release of the RRC session, this is interpreted by the UE as an implicit PTC deactivate command. At any time while in GAN mode, if the serving RR entity is switched to GSM-RR/UTRAN-RRC, the GA-RRC is disconnected from the GPRS SAPs and the UE enters GERAN/UTRAN mode. Simultaneously, the UE will release the associated PTC regardless of the PTC Timer status. The UE GA-RRC entity maintains one PTC for each active PDP context. The PTC Timer is restarted whenever any uplink user data packet is sent or downlink user data packet is received related to the PDP context. The PTC Timer value is provided to the UE as part of the GAN Registration procedure (i.e., in the GA-RC REGISTER ACCEPT message).

b) PTC Initial Activation

FIG. 85 illustrates the Packet Transport Channel initial activation procedure of some embodiments. The following description assumes theUE8505 is in the GA-RRC-IDLE state, in some embodiments. The GA-RRC Connection Establishment procedure is performed (in Step1) as described in clause UE Initiated GA-RRC Connection Establishment, above. TheUE8505 transitions to the GA-RRC-CONNECTED state and the PTC-STANDBY substate. Additional PS signaling procedures are performed (in Step2).

The CN8510 (SGSN) initiates (in Step3) the RAB Assignment procedure and includes the RAB-ID, the CN Transport Layer Address (IP address) and the CN Iu Transport Association (GTP-U Terminal Endpoint Identifier, TEID) for user data. TheGANC8515 sends (in Step4) the GA-RRC ACTIVATE PTC REQUEST message to theUE8505 to request activation of the Packet Transport Channel. The message includes the RAB-ID, and the CN IP Address and TEID to allow theUE8505 to send PTC packets (i.e., GTP-U messages) directly to the SGSN.

TheUE8505 acknowledges (in Step5) the PTC activation and provides the Transport Layer Address (IP address) and Iu Transport Association (GTP-U TEID) that identifies the UE end of the PTC. TheUE8505 transitions to the PTC-ACTIVE substate and starts the PTC Timer.

Upon receiving the acknowledgment, theGANC8515 sends (in Step6) the RAB Assignment Response message to the CN8510 (SGSN) to complete the RAB Assignment procedure and includes UE IP Address and GTP-U TEID. Additional PS signalling procedures are performed (in Step7); examples are illustrated in PDP Context Activation and Network Requested PDP Context Activation Sub-sections, below. TheUE8505 initiates (in Step8) uplink user data transfer via the established PTC and the CN8510 (SGSN) may use the same transport channel to send downlink user data packets.

c) PTC Data Transfer

FIG. 86 illustrates the transfer of GPRS user data packets via the GAN Packet Transport Channel in some embodiments. If required, the GAN PTC is established (in Step1) as specified in Sub-section VIII.B.9.b: “PTC Initial Activation”, above. Upon the GA-RRC PTC establishment, theUE8605 enters the PTC-ACTIVE substate and starts the PTC Timer. TheUE8605 initiates (in Step2) the transfer of an uplink user data packet using the standard GTP-U protocol as specified in 3GPP TS 29.060 and restarts the PTC Timer.

The CN8615 (SGSN) transfers (in Step3) downlink user data packet utilizing the same PTC associated with the specific PDP context. Downlink user data packets are transferred using the standard GTP-U protocol as specified in 3GPP TS 29.060. Upon receiving the downlink data packet, the UE restarts the associated PTC Timer. Additional uplink and downlink user data packets are transferred (in Step4) via the same PTC as described in

steps

2 and3, respectively. After each transmission/reception, theUE8605 restarts the PTC Timer.

d) UE initiated PTC Deactivation

FIG. 87 illustrates the scenario when the UE deactivates the Packet Transport Channel after the PTC Timer expires in some embodiments. TheUE8705 is (in Step1) in the GA-RRC-CONNECTED state and the PTC-ACTIVE substate. The PTC Timer associated with one of the active packet transport channels expires.

TheUE8705 sends (in Step2) the GA-RRC DEACTIVATE PTC REQUEST message to theGANC8710, including the RAB-ID to identify the PTC and indicating the normal release as a cause for deactivation. TheGANC8710 sends (in Step3) a RAB Release Request message to the CN (SGSN)8715 to request the release of the associated RAB. The CN (SGSN)8715 responds (in Step4) with the RAB Assignment Request indicating release.

TheGANC8710 responds (in Step5) to theUE8705 with a GA-RRC DEACTIVATE PTC ACK message to acknowledge successful deactivation. TheUE8705 transitions to the PTC-STANDBY substate. TheGANC8710 sends (in Step6) the RAB Assignment Response message to notify theSGSN8715 that the RAB Release procedure is complete.

E) UE Initiated PTC Re-Activation

FIG. 88 illustrates the scenario when the UE initiates re-activation of the Packet Transport Channel in some embodiments. The UE is in the GA-RRC-CONNECTED and PMM-CONNECTED states; e.g., a PS signaling connection and active PDP context exists between theUE8805 andCN8815 but the PTC was previously deactivated by theUE8805 due to PTC Timer expiry in some embodiments. TheUE8805 is in the GA-RRC-CONNECTED state and the PTC-STANDBY substate. TheUE8805 is in the PMM-CONNECTED state (i.e., a PS signaling connection and an active PDP context exists).

TheUE8805 has a PDU to send. TheUE8805 sends (in Step1) the Service Request message (with Service type value “Data”) to theGANC8810 in the GA-RRC UL DIRECT TRANSFER message. TheGANC8810 forwards (in Step2) the Service Request over the existing signaling connection to theCN8815 using the RANAP Direct Transfer message.

TheCN8815 may optionally initiate (in Step3) the Security Mode Control procedure described in Sub-section VIII.B.3: “Security Mode Control”, above. TheCN8815 responds (in Step4) with a Service Accept message. TheGANC8810 forwards (in Step5) the message to theUE8805.

TheUE8805,GANC8810 andCN8815 establish (in Step6) the GA-RRC Packet Transport Channel (PTC) as described in steps3-6 in VIII.B.9.b: “PTC Initial Activation”, above. TheUE8805 transitions to the PTC-ACTIVE substate and starts the PTC Timer. TheUE8805 sends (in Step7) the uplink PDU. Additional data transfer may take place.

F) Network Initiated PTC De-Activation

FIG. 89 illustrates the scenario when the network initiates de-activation of the Packet Transport Channel in some embodiments. TheUE8905 is in the GA-RRC-CONNECTED state and the PTC-ACTIVE substate.

Optionally, theGANC8910 may initiate the PTC de-activation procedure; e.g., as a result of an error handling procedure. If so, theGANC8910 sends (in Step1) the RAB Release Request message to theCN8915. The CN (SGSN)8915 sends (in Step2) a RAB Assignment Request to request the release of the associated RAB. The release request may include one or more RABs.

TheGANC8910 requests (in Step3) deactivation of the associated GA-RRC PTC by sending the GA-RRC DEACTIVATE PTC REQUEST message to theUE8905. TheUE8905 transitions to the PTC-STANDBY substate, stops the PTC Timer and sends (in Step4) the acknowledgment back to theGANC8910.

Steps

3 and4 are repeated for each additional RAB/PTC that needs to be released. TheGANC8910 notifies (in Step5) the CN (SGSN)8915 that the release was successful.

G) Network Initiated PTC Re-Activation

FIG. 90 illustrates the scenario when the network initiates re-activation of the Packet Transport Channel in some embodiments. TheUE9005 is in the GA-RRC-CONNECTED and PMM-CONNECTED states; e.g., a PS signaling connection and active PDP context exists between the UE and CN but the PTC was previously deactivated in some embodiments. TheUE9005 is in the GA-RRC-CONNECTED state and the PTC-STANDBY substate. TheUE9005 is in the PMM-CONNECTED state (i.e., a PS signaling connection and an active PDP context exists).

TheCN9015 has a PDU to send to send to theUE9005. TheCN9015 may optionally initiate (in Step1) the Security Mode Control procedure described in Sub-section VIII.B.3: “Security Mode Control”, above. TheUE9005,GANC9010 andCN9015 establish (in Step2) the GA-RRC Packet Transport Channel (PTC) as described in steps3-6 in clause in Sub-section VIII.B.9.b: “PTC Initial Activation”, above. TheUE9005 transitions to the PTC-ACTIVE substate and starts the PTC Timer. TheCN9015 sends (in Step3) the downlink PDU. Additional data transfer may take place.

H) Implicit PTC De-Activation Due to UE De-Registration

FIG. 96 illustrates the procedure for implicit PTC de-activation in some embodiments. As part of the GAN de-registration procedure, the GANC needs to release all resources allocated to thatUE9605. GAN de-registration may be initiated either explicitly by theUE9605 or implicitly by theGANC9610 if the loss of the signaling connection is detected. Initially, one or more GA-RRC PTCs associated with aUE9605 are in the PTC-ACTIVE state.

The GAN de-registration procedure is initiated (in Step1) for theUE9605 either by theUE9605 orGANC9610. Optionally, any outstanding resources associated with the CS Domain are released (in Step2). Optionally, if there are any outstanding resources associated with the PS Domain, theGANC9610 initiates (in Step3) the Iu release procedure to release the corresponding RABs. The CN (SGSN)9615 responds (in Step4) with Iu Release Command. Upon receiving the Iu Release Command, theGANC9610 locally deactivates (in Step5) all associated PTCs and responds (in Step6) to the core network (SGSN)9615 with an Iu Release Complete message.

10. PDP Context Activation

FIG. 91 illustrates the successful UE-initiated PDP Context Activation procedure, assuming the UE is in GA-RRC-IDLE mode in some embodiments. The GA-RRC Connection Establishment procedure is performed (in Step1) as described in Sub-section UE Initiated GA-RRC Connection Establishment, above. If a GA-RRC connection already exists (e.g., there is an existing CS call in progress), this step is skipped.

Upon request from the upper layers, theUE9105 sends (in Step2) the Service Request message (with Service type value “Signaling”) to theGANC9110 in the GA-RRC INITIAL DIRECT TRANSFER message. TheGANC9110 establishes an SCCP connection to theCN9115 and forwards (in Step3) the Service Request to theCN9115 using the RANAP Initial UE Message. Subsequent NAS messages between theUE9105 andcore network9115 will be sent betweenGANC9110 andCN9115 using the RANAP Direct Transfer message.

TheCN9115 may optionally authenticate (in Step4) theUE9105 using standard UTRAN authentication procedures. TheCN9115 may optionally initiate (in Step5) the Security Mode Control procedure described in Sub-section VIII.B.3: “Security Mode Control”, above.

The CN (SGSN)9115 responds (in Step6) with a Service Accept message. TheGANC9110 forwards (in Step6) the message to theUE9105. TheUE9105 sends (in Step7) the Activate PDP Context Request message providing details on the PDP context to theCN9115. This message is contained within the GA-RRC UL DIRECT TRANSFER between theUE9105 and theGANC9110. TheGANC9110 forwards (in Step7) the Activate PDP Context Request message to theCN9115.

TheUE9105,GANC9110 andCN9115 establish (in Step8) the GA-RRC Packet Transport Channel (PTC) as described in steps3-6 in Sub-section VIII.B.9.b: “PTC Initial Activation”, above. TheCN9115 indicates (in Step9) the PDP context establishment is complete using the Activate PDP Context Accept message to theGANC9110. GANC forwards (in Step9) this message to theUE9105 in the GA-RRC DL DIRECT TRANSFER message. TheUE9105 andCN9115 exchange (in Step10) user data transfer via the established PTC.

11. Network Requested PDP Context Activation

FIG. 92 illustrates the successful Network-Requested PDP Context Activation procedure, assuming the UE is in GA-RRC-IDLE mode, in some embodiments. Initially, the CN (SGSN)9215 received downlink user data to transfer to the UE and the associated RAB is not established. The UE is in PMM-IDLE state.

The CN (SGSN)9215 sends (in Step1) the RANAP Paging message to theUE9205 via theGANC9210 to locate the user. The paging request indicates paging for PS Domain signaling. TheGANC9210 forwards (in Step2) the paging information to theUE9205 in the GA-RRC PAGING REQUEST message.

TheUE9205 responds (in Step3) to theSGSN9215 via theGANC9210 with a Service Request message (with Service type value “Paging response”). The message is encapsulated within the GA-RRC INITIAL DIRECT TRANSFER message. TheGANC9210 forwards (in Step4) the Service Request message to theSGSN9215 encapsulated in the RANAP Initial UE Message.

TheCN9215 may optionally authenticate (in Step5) theUE9205 using standard UTRAN authentication procedures. TheCN9215 may optionally initiate (in Step6) the Security Mode Control procedure described in Sub-section VIII.B.3: “Security Mode Control”, above.

TheCN9215 sends (in Step7) the Request PDP Context Activation message to theGANC9210. TheGANC9210 forwards (in Step7) this message to theUE9205 in the GA-RRC DL DIRECT TRANSFER message.

TheUE9205 sends (in Step8) the Activate PDP Context Request message providing details on the PDP context to theCN9215. This message is contained within the GA-RRC UL DIRECT TRANSFER between theUE9205 and theGANC9210. The GANC forwards (in Step8) the Activate PDP Context Request message to theCN9215. TheUE9205,GANC9210 andCN9215 establish (in Step9) the GA-RRC Packet Transport Channel (PTC) as described in steps3-6 in Sub-section VIII.B.9.b: “PTC Initial Activation”, above.

TheCN9215 indicates (in Step10) the PDP context establishment is complete using the Activate PDP Context Accept message to theGANC9210. GANC forwards (in Step10) this message to theUE9205 in the GA-RRC DL DIRECT TRANSFER message. TheUE9205 andCN9215 exchange (in Step11) user data transfer via the established PTC.

12. PDP Context Activation with Active CS Session

FIG. 93 illustrates the successful UE-initiated PDP Context Activation procedure, assuming theUE9305 is in GA-RRC-CONNECTED mode (e.g., existing CS session) in some embodiments. The GA-RRC Connection Establishment procedure is performed as described in Sub-section UE Initiated GA-RRC Connection Establishment, above. If a GA-RRC connection already exists (e.g., there is an existing CS call in progress), this step is skipped.

Upon request from the upper layers, theUE9305 sends (in Step1) the Service Request message (with Service type value “Signaling”) to theGANC9310 in the GA-RRC INITIAL DIRECT TRANSFER message. TheGANC9310 establishes (in Step2) an SCCP connection to theCN9315 and forwards the Service Request to the CN using the RANAP Initial UE Message. Subsequent NAS messages between theUE9305 andcore network9315 will be sent betweenGANC9310 andCN9315 using the RANAP Direct Transfer message.

TheCN9315 may optionally authenticate (in Step3) theUE9305 using standard UTRAN authentication procedures. TheCN9315 may optionally initiate (in Step4) the Security Mode Control procedure described in Sub-section VIII.B.3: “Security Mode Control”, above.

The CN (SGSN)9315 responds (in Step5) with a Service Accept message. TheGANC9310 forwards (in Step5) the message to theUE9305. TheUE9305 sends (in Step6) the Activate PDP Context Request message providing details on the PDP context to theCN9315. This message is contained within the GA-RRC UL DIRECT TRANSFER between theUE9305 and theGANC9310. The GANC forwards (in Step6) the Activate PDP Context Request message to theCN9315.

TheUE9305,GANC9310 andCN9315 establish (in Step7) the GA-RRC Packet Transport Channel (PTC) as described in steps3-6 in Sub-section VIII.B.9.b: “PTC Initial Activation”, above. TheCN9315 indicates (in Step8) the PDP context establishment is complete using the Activate PDP Context Accept message to theGANC9310. GANC forwards (in Step8) this message to theUE9305 in the GA-RRC DL DIRECT TRANSFER message. TheUE9305 andCN9315 exchange (in Step9) user data transfer via the established PTC.

13. SRNS Relocation

Serving RNS Relocation Procedure is performed for an UE in PMM-CONNECTED state to move the RAN connection point from old RNC to the new RNC. Two scenarios will be considered: (1) SRNS Relocation from RNC to GANC; i.e from UTRAN to GAN, and (2) SRNS Relocation from GANC to RNC; i.e., from GAN to UTRAN. These procedures include several options based on the support for Iur interface and lossless SRNS Relocation. It is assumed in this version of the GAN Specification that the Iur interface is not supported. Additionally, given that PDCP protocol is not included in the GAN solution in order to optimize the data transport, it is assumed that the lossless SRNS Relocation is not supported either.

a) SRNS Relocation from UTRAN to GAN

FIG. 94 illustrates SRNS relocation procedure from UTRAN to GAN for a UE that is in PMM Connected state in some embodiments. It is assumed that Iur interface and lossless SRNS relocation procedure are not supported. Initially, theUE9405 is registered for GAN service and in PMM Connected state. At lease one PDP context is active with maximum bitrate higher than 0.

After detecting GAN coverage and successfully registering for GAN service, theUE9405 sends (in Step1) measurement report to theRNC9410 indicating the highest signal level for the GAN cell. TheRNC9410 sends (in Step2) Relocation Required message to the core network (SGSN)9420 to initiate the SRNS relocation procedure. The message indicates theGANC9415 as atarget RNC9410 and includes the information necessary for the relocation coordination.

The core network (SGSN)9420 forwards (in Step3) the request to theGANC9415. The message includes the list of the RABs that need to be setup and associated information. Based on the Relocation Request message, theCN9420 andGANC9415 establish (in Step4) requested RABs and associated PS Transport Channels as specified in GA-RRC Packet Transport Channel Management Procedures Sub-section, above.

TheGANC9415 responds (in Step5) to thecore network9420 with acknowledgment includingTarget RNC9410 to Source RNC Transport Container. The core network (SGSN)9420 proceeds (in Step6) with relocation by sending a Relocation Command to the old RNC that includes the Target RNC to Source RNC Transport Container.

TheRNC9410 starts forwarding (in Step7) of data to theUE9405 for the RABs that are subject to forwarding. The forwarding is performed for downlink user data only and is based on the Transport Layer Address and Iu Transport Association received from theGANC9415.

TheRNC9410 sends (in Step8) the PHYSICAL CHANNEL RECONFIGURATION message to theUE9405 to initiate relocation to GAN. TheRNC9410 continues with relocation by forwarding (in Step9) the SRNS Context information to theGANC9415 via the core network (SGSN)9420. The core network (SGNS)9420 forwards (in Step10) the SRNS Context to theGANC9415. TheGANC9415 responds (in Step11) with a Relocation Detect message.

TheUE9405 sends (in Step12) a GA-RRC Relocation Complete message to theGANC9415 to indicate successful relocation. TheGANC9415 sends (in Step13) the Relocation Complete message to the core network (SGSN)9420 to complete the procedure.

Upon receiving the Relocation Complete message, the core network (SGSN)9420 switches user plane fromRNC9410 to GANC (UE) and initiates (in Step14) Iu Release procedure towards theRNC9410. After the data forwarding timer expires and after releasing the associated resources, theRNC9410 responds (in Step15) with Iu Release Complete message to the core network (SGSN)9420.

14. Short Message Service

GAN provides support for both Circuit Switched and Packet Switched SMS services. GAN-attached and GPRS enabled UEs will be able to send and receive SMS messages via the GAN.

a) CS-Based SMS

CS-based SMS support in GAN is based on the same mechanism that is utilized for CS mobility management and call control. On the UE side, the SMS layers (including the supporting CM sub layer functions) utilize the services of the MM layer to transfer SMS messages per standard circuit switched UMTS implementation. The SM-CP protocol is effectively tunnelled between the UE and the CN, using GA-RRC messages from the UE to the GANC, where the GANC relays the SM-CP to RANAP messages for transport over the Iu-cs interface. As with the mobility management and call control procedures, the secure IPSec tunnel and TCP session are used to provide secure and reliable SMS delivery over the IP network.

b) PS-Based SMS

PS-based SMS message transfer is based on the same mechanism as the transfer of the PS mobility management and session management signaling messages. On the UE side, the SMS layers (including the supporting CM sub layer functions) utilize the services of the RRC (i.e., GA-RRC) layer to transfer SMS messages per standard packet switched UMTS implementation. As with mobility management and session management signaling, the secure IPsec tunnel and TCP session is used to provide secure and reliable PS-based SMS delivery over the IP network.

IX. COMPUTER SYSTEM

FIG. 95 conceptually illustrates a computer system with which some embodiments of the invention are implemented. Thecomputer system9500 includes abus9505, aprocessor9510, asystem memory9515, a read-only memory9520, apermanent storage device9525,input devices9530, andoutput devices9535.

Thebus9505 collectively represents all system, peripheral, and chipset buses that support communication among internal devices of thecomputer system9500. For instance, thebus9505 communicatively connects theprocessor9510 with the read-only memory9520, thesystem memory9515, and thepermanent storage device9525.

From these various memory units, theprocessor9510 retrieves instructions to execute and data to process in order to execute the processes of the invention. In some embodiments the processor comprises a Field Programmable Gate Array (FPGA), an ASIC, or various other electronic components for executing instructions. The read-only-memory (ROM)9520 stores static data and instructions that are needed by theprocessor9510 and other modules of the computer system. Thepermanent storage device9525, on the other hand, is a read-and-write memory device. This device is a non-volatile memory unit that stores instruction and data even when thecomputer system9500 is off. Some embodiments of the invention use a mass-storage device (such as a magnetic or optical disk and its corresponding disk drive) as thepermanent storage device9525. Some embodiments use one or more removable storage devices (flash memory card or memory stick) as the permanent storage device.

Like thepermanent storage device9525, thesystem memory9515 is a read-and-write memory device. However, unlikestorage device9525, the system memory is a volatile read-and-write memory, such as a random access memory. The system memory stores some of the instructions and data that the processor needs at runtime.

Instructions and/or data needed to perform processes of some embodiments are stored in thesystem memory9515, thepermanent storage device9525, the read-only memory9520, or any combination of the three. For example, the various memory units contain instructions for processing multimedia items in accordance with some embodiments. From these various memory units, theprocessor9510 retrieves instructions to execute and data to process in order to execute the processes of some embodiments.

Thebus9505 also connects to the input and

output devices

9530 and9535. The input devices enable the user to communicate information and select commands to the computer system. Theinput devices9530 include alphanumeric keyboards and cursor-controllers. Theoutput devices9535 display images generated by the computer system. The output devices include printers and display devices, such as cathode ray tubes (CRT) or liquid crystal displays (LCD). Finally, as shown inFIG. 95,bus9505 also couplescomputer9500 to anetwork9565 through a network adapter (not shown). In this manner, the computer can be a part of a network of computers (such as a local area network (“LAN”), a wide area network (“WAN”), or an Intranet) or a network of networks (such as the Internet).

It should be recognized by one of ordinary skill in the art that any or all of the components ofcomputer system9500 may be used in conjunction with the invention. For instance, some or all components of the computer system described with regards toFIG. 95 comprise some embodiments of the UE, FAP, GANC, and other equipments described above. Moreover, one of ordinary skill in the art will appreciate that any other system configuration may also be used in conjunction with the invention or components of the invention.

X. DEFINITIONS AND ABBREVIATIONS

The following is a list of definitions and abbreviations used:

- AAA Authentication, Authorization and Accounting
- AKA Authentication and Key Agreement
- AP Access Point
- AS Access Stratum
- BSC Base Station Controller
- BSS Base Station Subsystem
- BSSGP Base Station System GPRS Protocol
- BSSMAP Base Station System Management Application Part
- CC Call Control
- CGI Cell Global Identification
- CM Connection Management
- CN Core Network
- CS Circuit Switched
- CTM Cellular Text Telephone Modem
- DNS Domain Name System
- DTM Dual Transfer Mode
- EAP Extensible Authentication Protocol
- GA-CSR Generic Access—Circuit Switched Resources
- GA-PSR Generic Access—Packet Switched Resources
- GA-RC Generic Access—Resource Control
- GAN Generic Access Network
- GANC Generic Access Network Controller
- ETSI European Telecommunications Standards Institute
- FCC US Federal Communications Commission
- FQDN Fully Qualified Domain Name
- GAD Geographical Area Description
- GERAN GSM EDGE Radio Access Network
- GGSN Gateway GPRS Support Node
- GMM/SM GPRS Mobility Management and Session Management
- GPRS General Packet Radio Service
- GSM Global System for Mobile communications
- GSN GPRS Support Node
- HLR Home Location Register
- HPLMN Home PLMN
- IETF Internet Engineering Task Force
- IKE Internet Key Exchange
- IKEv2 IKE Version 2
- IMEISV International Mobile station Equipment Identity and Software Version number
- IMSI International Mobile Subscriber Identity
- IP Internet Protocol
- LA Location Area
- LAI Location Area Identity
- LLC Logical Link Control
- MAC Medium Access Control
- MAC Message Authentication Code
- MM Mobility Management
- MS Mobile Station
- MSC Mobile Switching Center
- MTP1 MessageTransfer Part layer 1
- MTP2 MessageTransfer Part layer 2
- MTP3 MessageTransfer Part layer 3
- NAS Non-Access Stratum
- PDP Packet Data Protocol
- PDU Protocol Data Unit
- PLMN Public Land Mobile Network
- PSAP Public Safety Answering Point—A PSAP is an emergency services network element that is responsible for answering emergency calls
- PSTN Public Switched Telephone Network
- P-TMSI Packet—TMSI
- QoS Quality of Service
- RA Routing Area
- RAC Routing Area Code
- RAI Routing Area Identity
- RAT Radio Access Technology
- RLC Radio Link Control
- RNC Radio Network Controller
- RNS Radio Network Subsystem
- RTCP Real Time Control Protocol
- RTP Real Time Protocol
- SCCP Signaling Connection Control Part
- SEGW SEcurity GateWay
- SGSN Serving GPRS Support Node
- SIM Subscriber Identity Module
- SMLC Serving Mobile Location Center
- SMS Short Message Service
- SNDCP Sub-Network Dependent Convergence Protocol
- TBF Temporary Block Flow
- TC Transport Channel
- TCP Transmission Control Protocol
- TFO Tandem Free Operation
- TMSI Temporary Mobile Subscriber Identity
- TrFO Transcoder Free Operation
- TTY Text Telephone or TeletYpewriter
- UE User Equipment
- UDP User Datagram Protocol
- UMTS Universal Mobile Telecommunication System
- UTRAN UMTS terrestrial Radio Access Network
- Up Up is the Interface between UE and GANC
- VLR Visited Location Register
- VPLMN Visited Public Land Mobile Network

While the invention has been described with reference to numerous specific details, one of ordinary skill in the art will recognize that the invention can be embodied in other specific forms without departing from the spirit of the invention. For instance the specific sequencing of procedures described and their associated attributes may be modified. Thus, one of ordinary skill in the art would understand that the invention is not to be limited by the foregoing illustrative details, but rather is to be defined by the appended claims.

Claims

1. A method of registering a user equipment (UE) in a communication system comprising a first licensed wireless communication system and a second generic access network (GAN) comprising a generic access network controller (GANC), the method comprising:

a) sending a register request message from the UE to the GANC indicating a GAN mode capability of A/Gb only for the UE;

b) when the GANC has a GAN mode capability of A/Gb, the GANC registering the UE with the GAN;

c) when the GANC has a GAN mode capability of Iu only, the GANC rejecting said register request message; and

d) when the GANC has a GAN mode capability of both A/Gb and Iu, the

GANC registering the UE based on a set of GANC mode selection rules for registering UEs with the GAN.

2. The method ofclaim 1, wherein said set of GANC mode selection rules utilizes information received by the GANC in the GAN register request message from the UE.

3. The method ofclaim 1, wherein said set of GANC mode selection rules utilizes information stored in the GANC.

4. The method ofclaim 1, wherein said set of GANC mode selection rules comprises service provider policies.