CN101132609B

Movatterモバイル変換

Info

Publication number: CN101132609B
Application number: CN2006101099128A
Authority: CN
Inventors: 邵飞; 郑潇潇
Original assignee: Huawei Technologies Co Ltd
Current assignee: Huawei Technologies Co Ltd
Priority date: 2006-08-22
Filing date: 2006-08-22
Publication date: 2010-04-21
Anticipated expiration: 2026-08-22
Also published as: WO2008025253A1; CN101132609A

Abstract

Translated fromChinese

本发明公开了一种切换过程中转发数据的方法及系统，用以解决目前切换过程中，下行数据转发会带来比较大的时延的问题。本发明方法包括：切换时，由源eNodeB按序将业务数据单元SDU转发到目标eNodeB，并告知转发数据的结束点；目标eNodeB按照接收顺序将收到的SDU向用户侧转发，并告知转发数据的结束点；用户侧RLC将收到的SDU递交到用户侧PDCP层，并告知转发数据的结束点；PDCP层以PDCP编号对转发数据结束点之前的SDU进行排序并向高层递交。本发明可以减少下行数据转发带来的时延。

The invention discloses a method and system for forwarding data in the switching process, which is used to solve the problem of relatively large time delay caused by downlink data forwarding in the current switching process. The method of the present invention includes: when switching, the source eNodeB forwards the service data unit SDUs to the target eNodeB in sequence, and informs the end point of the forwarded data; the target eNodeB forwards the received SDUs to the user side according to the receiving order, and informs the forwarded data The user-side RLC submits the received SDU to the user-side PDCP layer and informs the end point of the forwarded data; the PDCP layer sorts the SDUs before the end point of the forwarded data with the PDCP number and submits it to the upper layer. The present invention can reduce the delay caused by downlink data forwarding.

Description

Translated fromChinese

一种切换过程中转发数据的方法及系统A method and system for forwarding data during switching

技术领域technical field

本发明涉及通信领域，特别是涉及一种切换过程中转发数据的方法及系统。The present invention relates to the communication field, in particular to a method and system for forwarding data during switching.

背景技术Background technique

在20世纪90年代初开始了WCDMA的研究工作，从98年开始到现在，WCDMA系统的技术规范已经走过了Release 99，Release 4，Release 5，Release 6这几个阶段，目前关于Release 7的标准化工作已经开始实施。与此同时，更长期的LTE(Long Term Evolution)的研究也已经开始逐渐成为标准化工作的新热点。In the early 1990s, the WCDMA research work began. From 1998 to the present, the technical specifications of the WCDMA system have gone through the stages of Release 99, Release 4, Release 5, and Release 6. The current release 7 Standardization work has already started. At the same time, the longer-term LTE (Long Term Evolution) research has gradually become a new focus of standardization work.

在Release 99系统中，下行传输采用专用信道，其最大速率能够达到384kbps。从Release 4开始，将MSC(Mobile Service Switch Center)分裂为MSC Server和MGW(Multi-media GateWay)，从而将控制和业务分离。在Release 5中引入了IMS(IP Multi-media Subsystem)的概念，其基于IP的传输层，并在无线接入技术上引入了HSDPA(High Speed Downlink Packet Access)，使得下行速率可以达到14.4Mbps。在Release 6中引入了HSUPA(High SpeedUplink Packet Access)技术，使得上行速率能达到5.76Mbps。上述几个标准已经比较稳定，目前讨论的Release 7协议目标是在Release 6的基础上通过小改动，改善系统性能。LTE从系统框架到物理层都将是全新的，旨在为用户提供更高的速率，更好的性能服务。目前LTE中的UTRAN被称为E-UTRAN。In the Release 99 system, the downlink transmission adopts a dedicated channel, and its maximum rate can reach 384kbps. Starting from Release 4, the MSC (Mobile Service Switch Center) is split into MSC Server and MGW (Multi-media GateWay), thereby separating control and business. In Release 5, the concept of IMS (IP Multi-media Subsystem) was introduced, which is based on the IP transport layer, and HSDPA (High Speed Downlink Packet Access) was introduced in the wireless access technology, so that the downlink rate can reach 14.4Mbps. In Release 6, the HSUPA (High Speed Uplink Packet Access) technology was introduced, enabling the uplink rate to reach 5.76Mbps. The above-mentioned standards are relatively stable, and the goal of the Release 7 protocol currently being discussed is to improve system performance through minor changes on the basis of Release 6. LTE will be brand new from the system framework to the physical layer, aiming to provide users with higher speed and better performance services. The UTRAN in LTE is currently called E-UTRAN.

参见图1所示，为E-UTRAN系统架构图，其包括一个接入网关aGW(accessgateway)，以及与其通信的多个基站eNodeB。Referring to FIG. 1 , it is a diagram of an E-UTRAN system architecture, which includes an access gateway aGW (accessgateway) and multiple eNodeBs communicating with it.

所述aGW处于核心网，一个核心网中可包括多个aGW，aGW负责寻呼发起、LTE_IDLE的状态管理、用户面加密，分组数据汇聚协议PDCP(PacketData Convergence Protocol)、系统架构演进SAE(System Architecture Evolution)承载控制、非接入层NAS(Non Access stratum)信令的安全及完整性保护等功能。The aGW is in the core network, and a core network may include multiple aGWs. The aGW is responsible for paging initiation, LTE_IDLE state management, user plane encryption, packet data convergence protocol PDCP (PacketData Convergence Protocol), system architecture evolution SAE (System Architecture Evolution) bearer control, non-access stratum NAS (Non Access stratum) signaling security and integrity protection functions.

所述eNodeB位于接入网，与aGW相连，负责寻呼信息的调度及传输，广播信息的调度及传输，上下行资源的分配，无线承载控制，无线管理控制，LTE_ACTIVE状态下的连接移动性控制。The eNodeB is located in the access network and connected to the aGW, and is responsible for the scheduling and transmission of paging information, the scheduling and transmission of broadcast information, the allocation of uplink and downlink resources, radio bearer control, radio management control, and connection mobility control in LTE_ACTIVE state .

目前，E-UTRAN协议栈架构图，参见图2所示。At present, the architecture diagram of the E-UTRAN protocol stack is shown in FIG. 2 .

现有技术一：基于上述E-UTRAN协议栈架构，E-UTRAN系统中切换时的数据转发流程，参见图3所示。当源eNodeB收到切换准备确认后，开始向目标eNodeB转发数据，所述数据为RLC层的业务数据单元SDU，即PDCP层的协议数据单元PDU。在切换期间，RLC层进行重设，而从aGW向目标eNodeB发来的数据和源eNodeB转发来的数据都会存储在目标eNodeB中，等到目标eNodeB和用户终端同步上，并建立通信后，开始发送数据给用户终端。由于在切换时，RLC层发生重设，源eNodeB向目标eNodeB转发的数据可能与aGW向目标eNodeB发送的数据，交错到达目标eNodeB，所以目标eNodeB收到的源eNodeB转发来的数据可能是乱序的。Existing technology 1: Based on the above-mentioned E-UTRAN protocol stack architecture, the data forwarding process during handover in the E-UTRAN system is shown in FIG. 3 . When the source eNodeB receives the handover preparation confirmation, it starts to forward data to the target eNodeB, and the data is the service data unit SDU of the RLC layer, that is, the protocol data unit PDU of the PDCP layer. During the handover period, the RLC layer is reset, and the data sent from the aGW to the target eNodeB and the data forwarded by the source eNodeB will be stored in the target eNodeB. After the target eNodeB and the user terminal are synchronized and communication is established, the transmission will begin. data to the user terminal. Since the RLC layer is reset during the handover, the data forwarded by the source eNodeB to the target eNodeB may be interleaved with the data sent by the aGW to the target eNodeB to arrive at the target eNodeB, so the data forwarded by the source eNodeB received by the target eNodeB may be out of order of.

目标eNodeB的RLC层按接收顺序为所述乱序的数据重新编号后得到RLCPDU，但显然重新编号后得到RLC PDU仍然是乱序的。目标eNodeB再将重新编号后得到RLC PDU转发给用户终端，再由用户终端递交到PDCP层，显然无法保证业务的连贯性。The RLC layer of the target eNodeB renumbers the out-of-sequence data according to the received order to obtain RLC PDUs, but obviously the RLC PDUs obtained after renumbering are still out of order. The target eNodeB forwards the renumbered RLC PDU to the user terminal, and then the user terminal submits it to the PDCP layer, which obviously cannot guarantee the continuity of the service.

现有技术二：目前LTE跨不同eNodeB的切换流程，参见图4所示，包括下列步骤：Existing technology 2: the current LTE handover process across different eNodeBs, as shown in Figure 4, includes the following steps:

1、UE触发状态报告的原则由协议制定或系统消息下发等等。1. The principle for the UE to trigger the status report is formulated by the protocol or issued by the system message, etc.

2、源eNodeB根据UE上报的测量报告和RRM信息做切换决定。为了使目标eNodeB做切换准备，源eNodeB会将相关信息在切换请求消息中发送到目标eNodeB。2. The source eNodeB makes a handover decision according to the measurement report and RRM information reported by the UE. In order to make the target eNodeB prepare for handover, the source eNodeB will send relevant information to the target eNodeB in a handover request message.

3、目标eNodeB做层一层二的切换准备，并且向源eNodeB反馈新的小区无线网络临时标识(Cell Radio Network Temporary Identifier，C-RNTI)及可能的其他参数比如接入参数，SIB等等做为响应。在收到HO准备的确认后，源eNodeB开始向目标eNodeB转发数据包。3. The target eNodeB prepares for Layer 2 handover, and feeds back the new Cell Radio Network Temporary Identifier (C-RNTI) and possible other parameters such as access parameters, SIB, etc. to the source eNodeB. in response. After receiving the acknowledgment of HO preparation, the source eNodeB starts forwarding packets to the target eNodeB.

4、UE会收到含有新的C-RNTI以及可能的开始时间，目标e NB SIB等必须参数的HANDOVER COMMAND消息。这条RRC信息会用RLC AM模式传输，也就是说UE RLC会对这条信息进行确认以保证其可靠性。4. The UE will receive a HANDOVER COMMAND message containing necessary parameters such as the new C-RNTI, possible start time, and target e NB SIB. This RRC information will be transmitted in RLC AM mode, which means that UE RLC will confirm this information to ensure its reliability.

5、当HO COMMAND中的开始时间超时，UE将会执行与目标eNodeB的同步过程开始其上行同步。5. When the start time in HO COMMAND times out, the UE will perform the synchronization process with the target eNodeB to start its uplink synchronization.

6、网络进行上行资源分配和timing advance。这些资源是供UE发送切换确认“HANDOVER CONFIRM”消息给目标eNodeB用的，同时UE的切换过程结束。UE发送的“HO CONFIRM”消息是以RLC确认模式传输，需要网络侧的确认保证其可靠性。6. The network performs uplink resource allocation and timing advance. These resources are for the UE to send a handover confirmation "HANDOVER CONFIRM" message to the target eNodeB, and at the same time, the handover process of the UE ends. The "HO CONFIRM" message sent by the UE is transmitted in RLC confirmation mode, which requires confirmation from the network side to ensure its reliability.

7a、目标eNodeB通知源eNodeB成功切换，这样源eNodeB将删除缓存中已经转发的数据。如果源eNodeB缓存中还有数据或UPE仍继续向源eNodeB发送数据，源eNodeB会继续转发这些数据到目标eNodeB。7a. The target eNodeB notifies the source eNodeB of the successful handover, so that the source eNodeB deletes the forwarded data in the cache. If there is still data in the cache of the source eNodeB or the UPE still continues to send data to the source eNodeB, the source eNodeB will continue to forward these data to the target eNodeB.

7b、为了使UPE可以直接转发数据包到正确的目标eNodeB，UE位置信息变化会在移动性管理实体/数据面实体MME/UPE更新。7b. In order for the UPE to directly forward the data packet to the correct target eNodeB, the UE location information change will be updated in the mobility management entity/data plane entity MME/UPE.

LTE跨不同eNodeB切换时，第一种对用户数据的处理方案如下：When LTE is switched across different eNodeBs, the first processing scheme for user data is as follows:

1)当源eNodeB下发切换指令给UE(HO COMMAND)后，将源eNodeB中RLC缓存中的从第一个没有被UE正确确认的(即编号最小的PDU或SDU)开始所有SDU(根据编号)转发到目标eNodeB。1) After the source eNodeB sends the handover command to the UE (HO COMMAND), all SDUs (according to the number ) is forwarded to the target eNodeB.

2)目标eNodeB重建相应RLC实体，缓存转发来的数据包，当收到UE同步的信息后进行数据面传输。这样源eNodeB和目标eNodeB之间不需要交互RLC控制上下文(如SN等信息)，目标eNodeB将转发的数据进行重新编号或读取每个数据包高层编号后，向用户侧发送。2) The target eNodeB rebuilds the corresponding RLC entity, caches the forwarded data packets, and performs data plane transmission after receiving the UE synchronization information. In this way, there is no need to exchange RLC control context (such as SN and other information) between the source eNodeB and the target eNodeB. The target eNodeB renumbers the forwarded data or reads the high-level number of each data packet and sends it to the user side.

3)aGW在进行传输路径改变(path switch)后，高层数据包将会从aGW下发到目标eNodeB。这时会出现的问题，即第一个从aGW下发的新数据包(after path switching)到达目标eNodeB的时间，可能会比由源eNodeB转发的最后一个数据包到达目标eNodeB的时间早。这样在目标eNodeB侧就会出现数据包乱序到达的情况。为了解决这个问题，可以在eNodeB中，根据高层编号(PDCP编号)进行一定重排序(也可以不做)。3) After the aGW changes the transmission path (path switch), the high-level data packet will be delivered from the aGW to the target eNodeB. The problem that will arise at this time is that the time when the first new data packet (after path switching) delivered from the aGW reaches the target eNodeB may be earlier than the time when the last data packet forwarded by the source eNodeB arrives at the target eNodeB. In this way, data packets arrive out of order on the target eNodeB side. In order to solve this problem, a certain reordering can be performed (or not) in the eNodeB according to the higher layer number (PDCP number).

这种方案的问题在于转发从第一个没有被UE正确确认的数据包意味着一些不连续但已经被UE正确接收的数据会被重新转发，这就浪费了空口资源，也增加了切换时延(从高层角度看)。同时，这种大量的数据转发会造成从aGW高层发来的数据更晚到达目标eNodeB，若这种乱序如果在目标eNodeB进行重排序会带来比较大的时延。The problem with this solution is that forwarding the first data packet that has not been correctly acknowledged by the UE means that some discontinuous data that has been correctly received by the UE will be retransmitted, which wastes air interface resources and increases handover delay. (from a high-level perspective). At the same time, such a large amount of data forwarding will cause the data sent from the upper layer of the aGW to arrive at the target eNodeB later. If this disorder is reordered at the target eNodeB, it will bring a relatively large delay.

LTE跨不同eNodeB切换时，第二种对用户数据的处理方案如下：When LTE is switched across different eNodeBs, the second processing scheme for user data is as follows:

1)当源eNodeB下发切换指令给UE(HO COMMAND)后，将源eNodeB中RLC缓存中没有被UE正确确认的SDU(这些SDU可以是不连续的，选择性的)转发到目标eNodeB。1) After the source eNodeB sends the handover command to the UE (HO COMMAND), forward the SDUs in the RLC cache in the source eNodeB that are not correctly confirmed by the UE (these SDUs can be discontinuous and selective) to the target eNodeB.

2)目标eNodeB重建相应RLC实体，缓存转发的数据包，当收到UE同步的信息后进行数据面传输，对于这些转发的数据包进行重新编号或读取每个包高层的编号后进行发送。2) The target eNodeB rebuilds the corresponding RLC entity, caches the forwarded data packets, transmits the data plane after receiving the UE synchronization information, and renumbers these forwarded data packets or reads the number of each high-level layer before sending them.

3)这里同样也有乱序的问题。即aGW在进行传输路径改变(path switch)后，高层数据包将会从aGW下发到目标eNodeB，这时会出现第一个从aGW下发的数据包(after path switching)比由源eNodeB转发的最后一个数据包早到达目标eNodeB的情况，导致目标eNodeB出现数据包的乱序接收。3) There is also a problem of disorder here. That is, after the aGW changes the transmission path (path switching), the high-level data packet will be sent from the aGW to the target eNodeB. At this time, the first data packet sent from the aGW (after path switching) will be forwarded by the source eNodeB. The last data packet arrives at the target eNodeB early, causing the target eNodeB to receive out-of-order data packets.

这种进行选择性重传的方案相比第一种对用户数据的处理方案有所优化，不会浪费空口资源。但转发数据和传输路径转换后的数据发生乱序的问题仍无法解决，而且在目标eNodeB进行重排序会带来比较大的时延。This selective retransmission solution is optimized compared to the first solution for processing user data, and air interface resources will not be wasted. However, the problem of out-of-sequence of the forwarded data and the converted data of the transmission path still cannot be solved, and the reordering at the target eNodeB will bring relatively large time delay.

发明内容Contents of the invention

本发明提供一种切换过程中转发数据的方法及系统，用以解决目前切换过程中，下行数据转发会带来比较大的时延的问题。The present invention provides a method and system for forwarding data in the handover process, which is used to solve the problem of relatively large time delay caused by downlink data forwarding in the current handover process.

本发明方法包括下列步骤：The inventive method comprises the following steps:

A、切换时，由源eNodeB按序将业务数据单元SDU转发到目标eNodeB，并告知转发数据的结束点；A. During handover, the source eNodeB forwards the service data unit SDU to the target eNodeB in sequence, and informs the end point of the forwarded data;

B、目标eNodeB按照接收顺序将收到的SDU向用户侧转发，并告知转发数据的结束点；B. The target eNodeB forwards the received SDUs to the user side according to the receiving order, and informs the end point of the forwarded data;

C、用户侧RLC将收到的SDU递交到用户侧PDCP层，并告知转发数据的结束点；C. The RLC on the user side submits the received SDU to the PDCP layer on the user side, and informs the end point of the forwarded data;

D、PDCP层以PDCP编号对转发数据结束点之前的SDU进行排序并向高层递交。D. The PDCP layer uses the PDCP number to sort the SDUs before the end point of the forwarded data and submits them to the upper layer.

其中，步骤A中源eNodeB按序转发的SDU为没有被用户正确确认的SDU，或者为从第一个没有被用户正确确认的SDU。Wherein, the SDUs sequentially forwarded by the source eNodeB in step A are the SDUs that have not been correctly confirmed by the user, or the first SDUs that have not been correctly confirmed by the user.

其中，步骤A中所述告知转发数据结束点的方式为：为通过传输网络标识出转发的最后一个SDU；或者，在目标eNodeB侧设置定时器，若在定时时间内未收到一个SDU，则目标eNodeB据此判定转发数据结点到来。Wherein, the method of notifying the end point of the forwarded data in step A is: to identify the last SDU forwarded through the transmission network; or, set a timer on the target eNodeB side, if no SDU is received within the time limit, then Based on this, the target eNodeB determines the arrival of the forwarding data node.

其中，步骤B中，目标eNodeB将收到的SDU分割为协议数据单元PDU，并按照转发顺序对各个PDU进行RLC编号；步骤B中所述告知转发数据结束点的方式为：在发送的最后一个需要转发的SDU的最后一个PDU的包头中添加指示比特；或者，通过显示信令方式标识最后一个需要转发的SDU的最后一个PDU。Among them, in step B, the target eNodeB divides the received SDU into protocol data units (PDUs), and performs RLC numbering on each PDU according to the forwarding sequence; the method of notifying the end point of forwarded data in step B is: An indication bit is added to the header of the last PDU of the SDU to be forwarded; or, the last PDU of the last SDU to be forwarded is identified through explicit signaling.

进一步，步骤C中，所述用户侧RLC检测出本地缓存中存在完整的SDU，则立即向PDCP层递交。Further, in step C, the RLC at the user side detects that there is a complete SDU in the local cache, and immediately submits it to the PDCP layer.

步骤C中，用户侧RLC缓存目标eNodeB发来的PDU，并按照PDU的RLC编号排序，当用户侧RLC正确接收被标识的PDU后，对缓存的PDU进行SDU校验，之后将RLC编号小于等于被标识的PDU的RLC编号对应的SDU一并向PDCP层递交。In step C, the user-side RLC caches the PDUs sent by the target eNodeB, and sorts them according to the RLC numbers of the PDUs. When the user-side RLC correctly receives the identified PDUs, it performs SDU verification on the cached PDUs, and then sets the RLC numbers less than or equal to The SDU corresponding to the RLC number of the identified PDU is submitted to the PDCP layer together.

又进一步，步骤C中，当用户侧RLC将RLC编号小于等于被标识的PDU的RLC编号对应的SDU都递交到PDCP层或结束RLC排序时，通知PDCP层乱序接收结束，并按序递交后续收到的SDU。Still further, in step C, when the RLC on the user side submits the SDUs corresponding to the RLC numbers whose RLC number is less than or equal to the identified PDU to the PDCP layer or ends the RLC sorting, it notifies the PDCP layer that the out-of-order reception is completed, and submits the subsequent SDUs received.

更进一步，步骤B中，目标eNodeB将收到的SDU分割为协议数据单元PDU，并按照转发顺序对各个PDU进行RLC编号；步骤B中所述告知转发数据结束点的方式为：Furthermore, in step B, the target eNodeB divides the received SDU into protocol data units (PDUs), and performs RLC numbering on each PDU according to the forwarding order; the method of notifying the end point of the forwarded data as described in step B is:

目标eNodeB在转发的每个SDU的最后一个PDU的包头中添加指示比特，或通过显示信令方式标识转发的每个SDU的最后一个PDU。The target eNodeB adds an indication bit to the packet header of the last PDU of each SDU forwarded, or identifies the last PDU of each SDU forwarded through explicit signaling.

步骤C中，所述用户侧RLC检测出本地缓存中存在完整的SDU，则立即向PDCP层递交。In step C, the RLC at the user side detects that there is a complete SDU in the local cache, and immediately submits it to the PDCP layer.

步骤C中，所述用户侧RLC缓存目标eNodeB发来的PDU，并按照PDU的RLC编号排序，当用户侧RLC正确接收被标识的PDU后，对缓存的PDU进行SDU校验，之后将RLC编号小于等于被标识的PDU的RLC编号对应的SDU一并向PDCP层递交。In step C, the user-side RLC caches the PDUs sent by the target eNodeB, and sorts them according to the RLC numbers of the PDUs. When the user-side RLC correctly receives the identified PDUs, it performs SDU verification on the cached PDUs, and then the RLC numbers The SDU corresponding to the RLC number less than or equal to the identified PDU is submitted to the PDCP layer together.

再进一步，步骤C中，用户侧RLC将被标识的PDU对应的SDU都递交到PDCP层或结束RLC排序时，通知PDCP层乱序接收结束，并按序递交后续收到的SDU。Further, in step C, the user-side RLC submits all SDUs corresponding to the identified PDUs to the PDCP layer or completes the RLC sorting, notifies the PDCP layer of the end of out-of-order reception, and submits the subsequent received SDUs in sequence.

步骤B中，目标eNodeB将收到的SDU分割为PDU，并按照转发顺序对各个PDU进行RLC编号；步骤B中所述告知转发数据结束点的方式为：在发出最后一个需要转发的SDU的各个PDU之后，在从aGW传输来的第一个SDU的最后一个PDU的包头中添加指示比特；或者，通过显示信令方式标识从aGW传输来的第一个SDU的最后一个PDU。In step B, the target eNodeB divides the received SDU into PDUs, and performs RLC numbering on each PDU according to the forwarding order; the method of notifying the end point of the forwarded data in step B is: after sending out the last SDU that needs to be forwarded, each After the PDU, an indication bit is added to the header of the last PDU of the first SDU transmitted from the aGW; or, the last PDU of the first SDU transmitted from the aGW is identified through explicit signaling.

步骤C中，用户侧RLC缓存目标eNodeB发来的PDU，并按照PDU的RLC编号排序，当用户侧RLC收到被标识的PDU后，对缓存中RLC编号小于被标识的PDU的RLC编号的PDU进行SDU校验，之后将RLC编号小于被标识的PDU的RLC编号对应的SDU一并向PDCP层递交。In step C, the RLC on the user side caches the PDUs sent by the target eNodeB, and sorts them according to the RLC numbers of the PDUs. Perform SDU verification, and then submit the SDU corresponding to the RLC number whose RLC number is smaller than the identified PDU to the PDCP layer.

又进一步，步骤C中，用户侧RLC将RLC编号小于被标识的PDU的RLC编号对应的SDU都递交到PDCP层或结束RLC排序时，通知PDCP层乱序接收结束，并按序递交后续收到的SDU。Still further, in step C, the RLC on the user side submits the SDUs corresponding to the RLC numbers whose RLC number is smaller than the identified PDU to the PDCP layer or ends the RLC sorting, notifies the PDCP layer that the out-of-order reception is completed, and submits the subsequent received in order of SDUs.

更进一步，步骤B中，目标eNodeB将收到的SDU分割为PDU，并按照转发顺序对各个PDU进行RLC编号；步骤B中所述告知转发数据结束点的方式为：Furthermore, in step B, the target eNodeB divides the received SDU into PDUs, and performs RLC numbering on each PDU according to the forwarding sequence; the method of notifying the end point of forwarded data as described in step B is:

目标eNodeB在发出最后一个需要转发的SDU的各个PDU之后，在从aGW传输来的每个SDU的最后一个PDU的包头中添加指示比特，或者通过显示信令方式标识从aGW传输来的每个SDU的最后一个PDU。After sending out each PDU of the last SDU that needs to be forwarded, the target eNodeB adds an indication bit to the header of the last PDU of each SDU transmitted from the aGW, or identifies each SDU transmitted from the aGW through explicit signaling of the last PDU.

再进一步，当目标eNodeB收到任一一个被标识的PDU被正确接收的响应消息后，停止添加指示的操作。Still further, when the target eNodeB receives a response message indicating that any one of the identified PDUs has been received correctly, the operation of adding the indication is stopped.

进一步，步骤C中，所述用户侧RLC缓存目标eNodeB发来的PDU，并按照PDU的RLC编号排序，当用户侧RLC收到被标识的PDU后，对缓存中RLC编号小于被标识的PDU的RLC编号的PDU进行SDU校验，之后将RLC编号小于被标识的PDU的RLC编号对应的SDU一并向PDCP层递交。Further, in step C, the user-side RLC caches the PDUs sent by the target eNodeB, and sorts them according to the RLC numbers of the PDUs. The PDU with the RLC number performs SDU verification, and then submits the SDU corresponding to the RLC number with an RLC number smaller than the identified PDU to the PDCP layer.

步骤C中，用户侧RLC将RLC编号小于被标识的PDU的RLC编号对应的SDU都递交到PDCP层或结束RLC排序时，通知PDCP层乱序接收结束，并按序递交后续收到的SDU。In step C, the RLC on the user side submits the SDUs corresponding to the RLC numbers whose RLC number is smaller than the identified PDU to the PDCP layer or ends the RLC sorting, notifies the PDCP layer of the end of out-of-order reception, and submits the subsequent received SDUs in order.

步骤C中，用户侧RLC将校验后的SDU递交给PDCP层之后，通知PDCP层转发数据结束，并按序递交后续收到的SDU。In step C, after the RLC on the user side submits the verified SDU to the PDCP layer, it notifies the PDCP layer that data forwarding is over, and then submits the subsequently received SDUs in sequence.

又进一步，用户侧RLC将校验后的SDU递交给PDCP层之后，通知PDCP层转发数据结束，并按序递交后续收到的SDU。Still further, after the RLC at the user side submits the checked SDU to the PDCP layer, it notifies the PDCP layer that data forwarding is over, and then submits the subsequently received SDUs in sequence.

综上所述，所述PDCP层记录收到的PDU的PDCP编号与RLC编号的映射关系，根据该映射关系及转发数据结束点校验在转发数据结束点之前收到的PDU对应的SDU是否为乱序SDU。In summary, the PDCP layer records the mapping relationship between the PDCP number of the received PDU and the RLC number, and checks whether the SDU corresponding to the PDU received before the end point of the forwarding data is Out-of-sequence SDUs.

综上所述，所述用户侧RLC记录收到完整的并已经递交PDCP层的PDU的RLC编号，根据所述记录的PDU的RLC编号及转发数据结束点校验在转发数据结束点之前收到的PDU对应的SDU是否为乱序SDU。To sum up, the RLC on the user side records the RLC number of the PDU that has been received and has been submitted to the PDCP layer. According to the RLC number of the recorded PDU and the end point of the forwarded data, it is checked that the received data is received before the end point of the forwarded data. Whether the SDU corresponding to the PDU is an out-of-sequence SDU.

本发明系统，包括：The system of the present invention includes:

第一触发单元，其位于用户侧RLC层，用于在切换开始后收到目标eNodeB发来的第一个SDU时，触发所述递交单元；The first triggering unit is located at the RLC layer on the user side, and is used to trigger the handover unit when the first SDU sent by the target eNodeB is received after the start of the handover;

递交单元，其位于用户侧RLC层，用于向PDCP层递交SDU；A submitting unit, which is located at the RLC layer on the user side, and is used to submit the SDU to the PDCP layer;

第二触发单元，其位于PDCP层，用于在用户侧进行RLC重置时，触发排序单元开始排序。The second triggering unit is located at the PDCP layer, and is configured to trigger the sorting unit to start sorting when RLC reset is performed on the user side.

排序单元，其位于PDCP层，用于以PDCP编号对转发数据结束点之前递交单元发来的SDU进行排序；A sorting unit, which is located at the PDCP layer, is used to sort the SDUs sent by the submission unit before the end point of forwarding data by PDCP number;

转发数据结束点通知单元，其位于源eNodeB、目标eNodeB及用户侧RLC中，用于通知转发数据结束点。The forwarding data end point notification unit is located in the source eNodeB, the target eNodeB and the user side RLC, and is used for notifying the forwarding data end point.

本发明有益效果如下：The beneficial effects of the present invention are as follows:

本发明方法不在RLC层进行乱序重排序，而是在切换时，由源eNodeB依次通过目标eNodeB及用户侧RLC，将没有被用户正确确认的SDU或从第一个没有被用户正确确认的SDU递交到用户侧的PDCP层，并依次告知转发数据的结束点；PDCP层以PDCP编号对转发数据的结束点之前的SDU进行排序并向高层递交。The method of the present invention does not perform out-of-order reordering at the RLC layer, but when switching, the source eNodeB passes through the target eNodeB and the user-side RLC in sequence, and the SDUs that have not been correctly confirmed by the user or the first SDU that has not been correctly confirmed by the user Submit to the PDCP layer on the user side, and inform the end point of the forwarded data in turn; the PDCP layer sorts the SDUs before the end point of the forwarded data with the PDCP number and submits it to the upper layer.

为了支撑本发明方法，本发明还提供了一种切换过程中转发数据的系统。In order to support the method of the present invention, the present invention also provides a system for forwarding data during handover.

综上所述，本发明在用户侧的PDCP层进行重排序可以支持在目标eNodeB不需要重排序，从而减少下行数据转发带来的时延。In summary, the present invention performs reordering at the PDCP layer on the user side to support no reordering at the target eNodeB, thereby reducing the time delay caused by downlink data forwarding.

在用户侧的PDCP层重排序，相对于现有用户侧RLC顺序递交方案的可靠性有了更严格的保证，减少更高层处理的复杂度。Compared with the reliability of the existing user-side RLC sequential delivery scheme, the reordering at the PDCP layer on the user side has a stricter guarantee and reduces the complexity of higher-layer processing.

用户侧的PDCP层排序时，用户侧RLC可以不进行PDU的RLC编号排序，即可以支持乱序递交，从而减少了用户侧不必要的两层排序带来的处理时延。When sorting at the PDCP layer on the user side, the RLC on the user side does not need to sort the RLC numbers of the PDUs, that is, it can support out-of-order delivery, thereby reducing the processing delay caused by unnecessary two-layer sorting on the user side.

附图说明Description of drawings

图1为E-UTRAN系统架构图；Figure 1 is a diagram of the E-UTRAN system architecture;

图2为E-UTRAN协议栈架构图；Fig. 2 is the architecture diagram of the E-UTRAN protocol stack;

图3为E-UTRAN系统中切换时的数据转发流程图；Figure 3 is a flow chart of data forwarding during handover in the E-UTRAN system;

图4为目前LTE跨不同eNodeB的切换流程图；Figure 4 is a flow chart of current LTE handover across different eNodeBs;

图5为本发明方法步骤流程图；Fig. 5 is a flowchart of the method steps of the present invention;

图6为本发明系统结构示意图。Fig. 6 is a schematic diagram of the system structure of the present invention.

具体实施方式Detailed ways

为了在切换过程中，减少下行数据转发带来的时延，本发明提供一种切换过程中转发数据的方法，参见图5所示，包括下列主要步骤：In order to reduce the delay caused by downlink data forwarding during the handover process, the present invention provides a method for forwarding data during the handover process, as shown in FIG. 5 , including the following main steps:

S1、切换时，由源eNodeB按序将SDU转发到目标eNodeB，并告知转发数据的结束点。S1. During handover, the source eNodeB forwards the SDUs to the target eNodeB in sequence, and notifies the end point of the forwarded data.

切换时，源eNodeB将没有被用户正确确认的SDU或从第一个没有被用户正确确认的SDU按照接收顺序(即按序转发)转发给目标eNodeB，并告知转发数据的结束点。During handover, the source eNodeB forwards the SDUs that are not correctly confirmed by the user or from the first SDU that is not correctly confirmed by the user to the target eNodeB in the order of receipt (that is, sequential forwarding), and informs the end point of the forwarded data.

源eNodeB在向目标eNodeB转发SDU的过程中，通过传输网络标识出转发的最后一个SDU(即为转发的最后一个SDU添加标签，以指示乱序接收的结束点，并通过FP传输)；或者，在目标eNodeB侧设置定时器，若在定时时间内未收到一个SDU，则目标eNodeB据此判定转发数据结点到来。In the process of forwarding SDUs to the target eNodeB, the source eNodeB identifies the last SDU forwarded through the transmission network (that is, adds a label to the last SDU forwarded to indicate the end point of out-of-order reception, and transmits it through FP); or, A timer is set on the target eNodeB side. If an SDU is not received within the time limit, the target eNodeB will judge the arrival of the forwarding data node accordingly.

S2、目标eNodeB按照接收顺序将收到的SDU向用户侧转发，并告知转发数据的结束点。S2. The target eNodeB forwards the received SDUs to the user side according to the receiving sequence, and notifies the end point of the forwarded data.

本步骤中，目标eNodeB将收到的SDU分割为PDU，并按照转发顺序对各个PDU进行RLC编号；基于分割的PDU，目标eNodeB告知用户侧转发数据结束点的方式为下列方式之一：In this step, the target eNodeB divides the received SDU into PDUs, and performs RLC numbering on each PDU according to the forwarding order; based on the divided PDUs, the target eNodeB informs the user side of the end point of forwarding data in one of the following ways:

方式11：目标eNodeB在发送的最后一个需要转发的SDU的最后一个PDU的包头中添加指示比特；或者，通过显示信令方式标识最后一个需要转发的SDU的最后一个PDU。Mode 11: The target eNodeB adds an indication bit to the packet header of the last PDU of the last SDU to be forwarded; or, identifies the last PDU of the last SDU to be forwarded through explicit signaling.

方式12：(作为方式一的变形方式)目标eNodeB在转发的每个SDU的最后一个PDU的包头中添加指示比特；或者，通过显示信令方式标识转发的每个SDU的最后一个PDU。即不仅最后一个需要转发的SDU的最后一个PDU被标识，而且其之前的每个SDU的最后一个PDU都被标识。Mode 12: (as a variant of Mode 1) the target eNodeB adds an indication bit to the packet header of the last PDU of each SDU forwarded; or, identifies the last PDU of each SDU forwarded by means of explicit signaling. That is, not only the last PDU of the last SDU to be forwarded is identified, but also the last PDU of each SDU before it is identified.

方式13：目标eNodeB在发出最后一个需要转发的SDU的各个PDU之后，在从aGW传输来的第一个SDU的最后一个PDU的包头中添加指示比特；或者，通过显示信令方式标识从aGW传输来的第一个SDU的最后一个PDU。Method 13: After the target eNodeB sends out each PDU of the last SDU to be forwarded, add an indication bit to the header of the last PDU of the first SDU transmitted from the aGW; or, identify the transmission from the aGW by means of explicit signaling The last PDU of the first SDU to come.

方式14：(作为方式三的变形方式)目标eNodeB在发出最后一个需要转发的SDU的各个PDU之后，在从aGW传输来的每个SDU的最后一个PDU的包头中添加指示比特；或者，通过显示信令方式标识从aGW传输来的每个SDU的最后一个PDU。当目标eNodeB收到任一一个被标识的PDU被正确接收的响应消息后，停止添加指示的操作。Mode 14: (as a variant of Mode 3) after sending out each PDU of the last SDU to be forwarded, the target eNodeB adds an indication bit to the packet header of the last PDU of each SDU transmitted from the aGW; or, by displaying The signaling method identifies the last PDU of each SDU transmitted from the aGW. When the target eNodeB receives a response message indicating that any one of the identified PDUs has been received correctly, it stops the operation of adding indications.

S3、用户侧RLC将收到的SDU递交到用户侧PDCP层，并告知转发数据的结束点。S3. The RLC at the user side submits the received SDU to the PDCP layer at the user side, and notifies the end point of the forwarded data.

用户侧RLC向PDCP层乱序递交的功能，在切换开始后收到目标eNodeB发来的第一个SDU时触发。The out-of-sequence delivery function of the RLC on the user side to the PDCP layer is triggered when the first SDU sent by the target eNodeB is received after the start of the handover.

本步骤中，用户侧将SDU递交到PDCP层的方式为下列方式之一：In this step, the user side submits the SDU to the PDCP layer in one of the following ways:

方式21(对应方式11至方式14)：用户侧RLC检测出本地缓存中存在完整的SDU，则立即向PDCP层递交(乱序递交)。Mode 21 (corresponding to Mode 11 to Mode 14): the RLC at the user side detects that there is a complete SDU in the local cache, and immediately submits it to the PDCP layer (out-of-order delivery).

方式22(对应方式11)：用户侧RLC缓存目标eNodeB发来的PDU，并按照PDU的RLC编号排序，当用户侧RLC正确接收被标识的PDU后，对缓存的PDU进行SDU校验，之后将RLC编号小于等于被标识的PDU的RLC编号对应的SDU一并向PDCP层递交。Mode 22 (corresponding to Mode 11): The RLC on the user side caches the PDUs sent by the target eNodeB and sorts them according to the RLC numbers of the PDUs. The SDU corresponding to the RLC number whose RLC number is less than or equal to the identified PDU is submitted to the PDCP layer together.

方式23(对应方式12)：用户侧RLC缓存目标eNodeB发来的PDU，并按照PDU的RLC编号排序，当用户侧RLC收到未被标识的PDU后，对缓存中RLC编号小于未被标识的PDU的RLC编号的PDU进行SDU校验，之后将RLC编号小于未被标识的PDU的RLC编号对应的SDU一并向PDCP层递交。Mode 23 (corresponding to Mode 12): The RLC on the user side caches the PDUs sent by the target eNodeB and sorts them according to the RLC numbers of the PDUs. The PDU with the RLC number of the PDU performs SDU verification, and then submits the SDU corresponding to the RLC number of the PDU whose RLC number is smaller than that of the unidentified PDU to the PDCP layer.

方式24(对应方式13和方式14)：Mode 24 (corresponding to Mode 13 and Mode 14):

用户侧RLC缓存目标eNodeB发来的PDU，并按照PDU的RLC编号排序，当用户侧RLC收到被标识的PDU后，对缓存中RLC编号小于被标识的PDU的RLC编号的PDU进行SDU校验，之后将RLC编号小于被标识的PDU的RLC编号对应的SDU一并向PDCP层递交。The RLC on the user side caches the PDUs sent by the target eNodeB and sorts them according to the RLC numbers of the PDUs. When the RLC on the user side receives the identified PDUs, it performs SDU verification on the PDUs whose RLC numbers in the cache are smaller than the RLC numbers of the identified PDUs. , and then submit the SDU corresponding to the RLC number whose RLC number is smaller than the identified PDU to the PDCP layer.

本步骤中，用户侧告知PDCP层转发数据的结束点的方式为下列方式之一：In this step, the user side notifies the PDCP layer of the end point of forwarding data in one of the following ways:

方式31(对应上述方式21+方式11)：当用户侧RLC将RLC编号小于等于被标识的PDU的RLC编号对应的SDU都递交到PDCP层或结束RLC排序时，通知PDCP层乱序接收结束，并按序递交后续收到的SDU。Mode 31 (corresponding to the above mode 21 + mode 11): When the RLC on the user side submits the SDUs corresponding to the RLC numbers whose RLC numbers are less than or equal to the identified PDUs to the PDCP layer or ends the RLC sorting, it notifies the PDCP layer that the out-of-sequence reception is over, And submit the subsequent received SDUs in sequence.

方式32(对应上述方式21+方式12)：用户侧RLC将被标识的PDU对应的SDU都递交到PDCP层或结束RLC排序时，通知PDCP层乱序接收结束，并按序递交后续收到的SDU。Mode 32 (corresponding to the above mode 21 + mode 12): when the RLC on the user side submits all the SDUs corresponding to the identified PDUs to the PDCP layer or ends the RLC sorting, it notifies the PDCP layer that the out-of-order reception is over, and submits the subsequent received SDUs in order SDUs.

方式33：(对应上述方式21+方式13；方式21+方式14)：用户侧RLC将RLC编号小于被标识的PDU的RLC编号对应的SDU都递交到PDCP层或结束RLC排序时，通知PDCP层乱序接收结束，并按序递交后续收到的SDU。Mode 33: (corresponding to the above mode 21 + mode 13; mode 21 + mode 14): the RLC on the user side notifies the PDCP layer when it submits all SDUs corresponding to the RLC numbers whose RLC numbers are smaller than the identified PDUs to the PDCP layer or ends the RLC sorting The out-of-order reception ends, and the subsequent received SDUs are delivered in order.

方式34(对应上述方式22、方式23和方式24)：用户侧RLC将校验后的SDU递交给PDCP层之后，通知PDCP层转发数据结束，并按序递交后续收到的SDU。Mode 34 (corresponding to Mode 22, Mode 23, and Mode 24 above): After the user-side RLC submits the verified SDU to the PDCP layer, it notifies the PDCP layer of the end of data forwarding, and submits the subsequent received SDUs in sequence.

进一步，基于上述用户侧将SDU递交到PDCP层的方式，以及与之对应的转发数据结束点，用户侧RLC还可记录收到完整的并已经递交PDCP层的PDU的RLC编号，根据所述记录的PDU的RLC编号及转发数据结束点校验在转发数据结束点之前收到的PDU对应的SDU是否为乱序SDU，从而避免后续顺序的SUD先于最后一个乱序SDU到达用户侧，导致顺序SUD被误认为乱序SDU，进而被递交到PDCP层进行误排序的问题(也可在步骤S4中进行)。Further, based on the above-mentioned way that the user side submits the SDU to the PDCP layer, and the corresponding end point of forwarding data, the user side RLC can also record the RLC number of the PDU that has received the complete PDU that has been submitted to the PDCP layer. According to the record The RLC number of the PDU and the end point of the forwarded data check whether the SDU corresponding to the PDU received before the end point of the forwarded data is an out-of-order SDU, so as to prevent the subsequent SDUs from arriving at the user side before the last out-of-order SDU, resulting in a sequence The SUD is mistaken as an out-of-sequence SDU, and then submitted to the PDCP layer for misordering (also in step S4).

S4、PDCP层以PDCP编号对转发数据结束点之前的SDU进行排序并向高层递交。S4. The PDCP layer sorts the SDUs before the end point of the forwarded data by the PDCP number and submits them to the upper layer.

PDCP层在用户侧进行RLC重置时触发排序功能，PDCP层提取收到的每一PDU的PDCP编号，并以PDCP编号对SDU重排序，直至收到用户侧发来的转发数据结束通知时，结束排序，并将排序后的SDU向更高层递交。之后按接收顺序向更高层递交SDU。The PDCP layer triggers the sorting function when the RLC is reset on the user side. The PDCP layer extracts the PDCP number of each PDU received, and reorders the SDUs with the PDCP number until it receives the notification of the end of forwarded data from the user side. End the sorting, and submit the sorted SDUs to higher layers. The SDUs are then delivered to higher layers in the order they were received.

进一步，PDCP层也可记录收到的PDU的PDCP编号与RLC编号的映射关系，根据该映射关系及转发数据结束点校验在转发数据结束点之前收到的PDU对应的SDU是否为乱序SDU。Further, the PDCP layer can also record the mapping relationship between the PDCP number of the received PDU and the RLC number, and check whether the SDU corresponding to the PDU received before the end point of the forwarded data is an out-of-sequence SDU according to the mapping relationship and the end point of the forwarded data .

以下通过八个实例具体描述本发明方法。The method of the present invention is specifically described below through eight examples.

方法实例一：上述方式11、方式21和方式31的组合。Method example 1: a combination of the above-mentioned method 11, method 21 and method 31.

L11、当源eNodeB收到目标eNodeB的切换确认或源eNodeB向用户侧发送切换命令后，源eNodeB按照其接收顺序，将没有被用户侧正确确认的SDU或者从第一个没有被用户正确确认的SDU向目标eNodeB转发。同时通过传输网络标识出转发的最后一个SDU；或者，在目标eNodeB侧设置定时器，若在定时时间内未收到一个SDU，则目标eNodeB据此判定转发数据结点到来。L11. When the source eNodeB receives the handover confirmation from the target eNodeB or the source eNodeB sends a handover command to the user side, the source eNodeB will, according to the order in which it is received, select the SDUs that are not correctly confirmed by the user side or start from the first SDU that is not correctly confirmed by the user side The SDU is forwarded to the target eNodeB. At the same time, the last SDU forwarded is identified through the transmission network; or, a timer is set on the target eNodeB side, and if an SDU is not received within the time limit, the target eNodeB determines the arrival of the forwarding data node accordingly.

L12、目标eNodeB生成相应的RLC实体，以缓存收到的SDU。当收到用户侧的切换确认或同步信息后，开始用户面数据的发送。在转发SDU的过程中，目标eNodeB按照SDU数据包到达的先后，顺序对收到的SDU重新进行PDU的RLC编号(即将SDU分割为若干个PDU，并顺序对所述PDU编号)，或者读取SDU中PDCP编号，再按照接收顺序向用户侧发送。L12. The target eNodeB generates a corresponding RLC entity to buffer the received SDU. After receiving the handover confirmation or synchronization information from the user side, the user plane data transmission starts. In the process of forwarding SDUs, the target eNodeB re-numbers the received SDUs according to the sequence of arrival of the SDU data packets (that is, divides the SDU into several PDUs and sequentially numbers the PDUs), or reads The PDCP number in the SDU is sent to the user side in the order in which it is received.

本步骤中，在目标eNodeB侧不以PDCP编号对收到的SDU排序。In this step, the received SDUs are not sorted by PDCP numbers at the target eNodeB side.

L13、目标eNodeB在向用户侧乱序发送SDU的过程中，在发送的最后一个需要转发的SDU的最后一个PDU的包头中添加指示比特，或者通过显示信令方式标识最后一个需要转发的SDU的最后一个PDU，以告知用户侧转发数据结束点。L13. In the process of sending SDUs out of sequence to the user side, the target eNodeB adds an indication bit to the packet header of the last PDU of the last SDU that needs to be forwarded, or identifies the last SDU that needs to be forwarded by means of explicit signaling The last PDU is used to inform the user side of the end point of forwarding data.

L14、当用户侧收到切换命令消息后(可上接步骤L11)，开始进行重置RLC实体的操作，包括：通知高层(PDCP层)生成(激活)PDCP排序单元，开始乱序排序；检测RLC接收缓存中是否存在完整的SDU；删除不完整的SDU分段；初始化用户侧记时器和状态变量等参数。L14, after the user side receives the handover command message (can be connected to step L11), start the operation of resetting the RLC entity, including: notify the high layer (PDCP layer) to generate (activate) the PDCP sorting unit, start out of order sorting; detect Whether there is a complete SDU in the RLC receiving buffer; delete incomplete SDU segments; initialize parameters such as user-side timers and state variables.

L15、对于重置后的用户侧RLC实体，收到目标eNodeB发来的第一个SDU时，触发RLC接收缓存重组乱序递交功能，只要检测出完整的SDU，则立即向PDCP层递交，不管PDU的RLC编号的先后。用户侧开始以乱序递交的方式将从目标eNodeB接收到的SDU，向PDCP层递交。L15. For the user-side RLC entity after reset, when the first SDU sent by the target eNodeB is received, the RLC reception buffer reorganization and out-of-order delivery function is triggered. As long as a complete SDU is detected, it is immediately submitted to the PDCP layer, regardless of The order of the RLC numbers of the PDUs. The user side starts to deliver the SDU received from the target eNodeB to the PDCP layer in an out-of-order delivery manner.

L16、当用户侧RLC将RLC编号小于等于被标识的PDU的RLC编号对应的SDU都递交到PDCP层或结束RLC排序时，通知PDCP层乱序接收结束，并按序递交后续收到的SDU；即正确接收被标识的PDU后，用户侧对RLC接收缓存中PDU的RLC编号大于被标识的PDU的RLC编号N对应的SDU实行按序递交，并且必须等到PDU的RLC编号小于等于N的PDU被成功接收或被删除后，大于N的PDU对应的完整SDU才可以重组送到PDCP层。所述PDU被删除是根据设定重排序机制，由窗口或记时器控制。L16. When the RLC on the user side submits the SDUs corresponding to the RLC numbers whose RLC number is less than or equal to the identified PDU to the PDCP layer or ends the RLC sorting, notify the PDCP layer that the out-of-order reception is completed, and submit the subsequent received SDUs in order; That is, after receiving the identified PDU correctly, the user side will deliver the SDUs corresponding to the RLC number N of the PDU in the RLC receive buffer in sequence, and must wait until the PDU whose RLC number is less than or equal to N is received. After being successfully received or deleted, the complete SDU corresponding to the PDU greater than N can be reassembled and sent to the PDCP layer. The PDU is deleted according to a set reordering mechanism, controlled by a window or a timer.

L17、PDCP层根据收到的SDU的PDCP编号进行重排序。L17. The PDCP layer performs reordering according to the PDCP numbers of the received SDUs.

L18、当PDU的RLC编号小于等于N的每一PDU都被递交到PDCP层或达到最大重传次数或移出接收窗口时，用户侧RLC实体通知PDCP实体，PDCP将排序后的SDU送到更高层，挂起其排序功能。L18. When each PDU whose RLC number is less than or equal to N is delivered to the PDCP layer or reaches the maximum number of retransmissions or moves out of the receiving window, the user-side RLC entity notifies the PDCP entity, and PDCP sends the sorted SDUs to a higher layer , suspending its sort function.

方法实例二：上述方式12、方式21和方式32的组合。Method example 2: a combination of the above-mentioned method 12, method 21 and method 32.

L21和L22与L11和L12相同。L21 and L22 are the same as L11 and L12.

L23、目标eNodeB在向用户侧发送SDU的过程中，在转发的每个SDU的最后一个PDU的包头中添加指示比特；或者，通过显示信令方式标识转发的每个SDU的最后一个PDU。L23. In the process of sending the SDU to the user side, the target eNodeB adds an indication bit to the packet header of the last PDU of each SDU forwarded; or identifies the last PDU of each SDU forwarded by explicit signaling.

L24与L14相同。L24 is the same as L14.

L25与L15相同。L25 is the same as L15.

L26、用户侧RLC将被标识的PDU对应的SDU都递交到PDCP层或结束RLC排序时，通知PDCP层乱序接收结束，并按序递交后续收到的SDU；即正确接收未被标识的PDU后，用户侧对RLC接收缓存中PDU的RLC编号大于等于未被标识的PDU的RLC编号N对应的SDU实行按序递交，并且必须等到PDU的RLC编号小于N的PDU被成功接收或被删除后，大于等于N的PDU对应的完整SDU才可以重组送到PDCP层。所述PDU被删除是根据设定重排序机制，由窗口或记时器控制。L26. When the RLC on the user side submits the SDUs corresponding to the identified PDUs to the PDCP layer or ends the RLC sorting, it notifies the PDCP layer that the out-of-order reception is completed, and submits the subsequent received SDUs in order; that is, the unidentified PDUs are correctly received. Afterwards, the user side performs sequential delivery of the SDUs corresponding to the RLC number N of the PDU in the RLC reception buffer that is greater than or equal to the RLC number N of the unidentified PDU, and must wait until the PDU whose RLC number is less than N is successfully received or deleted , the complete SDU corresponding to the PDU greater than or equal to N can be reassembled and sent to the PDCP layer. The PDU is deleted according to a set reordering mechanism, controlled by a window or a timer.

L27与L17相同。L27 is the same as L17.

L28、当PDU的RLC编号小于N的每一PDU都被递交到PDCP层或达到最大重传次数或移出接收窗口时，用户侧RLC实体通知PDCP实体，PDCP将排序后的SDU送到更高层，挂起其排序功能。L28. When each PDU whose RLC number of the PDU is less than N is delivered to the PDCP layer or reaches the maximum number of retransmissions or moves out of the receiving window, the user-side RLC entity notifies the PDCP entity, and PDCP sends the sorted SDUs to a higher layer, Suspend its sort function.

方法实例三：上述方式11、方式22和方式34的组合。Method Example 3: Combination of the above method 11, method 22 and method 34.

L31至L34与L11至L14相同。L31 to L34 are the same as L11 to L14.

L35、用户侧RLC缓存目标eNodeB发来的PDU，并按照PDU的RLC编号排序，当用户侧RLC正确接收被标识的PDU后，对缓存的PDU进行SDU校验(即对RLC编号小于等于被标识的PDU的RLC编号N的所有PDU进行SDU检测)，之后将RLC编号小于等于被标识的PDU的RLC编号对应的SDU一并向PDCP层递交。对编号大于N的PDU对应的SDU按序递交。L35. The user-side RLC caches the PDUs sent by the target eNodeB, and sorts them according to the RLC numbers of the PDUs. When the user-side RLC correctly receives the identified PDUs, it performs SDU verification on the cached PDUs (that is, the RLC numbers are less than or equal to the identified PDUs). All PDUs with the RLC number N of the PDU of the PDU are subjected to SDU detection), and then the SDUs with the RLC number less than or equal to the RLC number of the identified PDU are submitted to the PDCP layer together. SDUs corresponding to PDUs with numbers greater than N are delivered sequentially.

L36、用户侧RLC将校验后的SDU递交给PDCP层之后，通知PDCP层转发数据结束，并按序递交后续收到的SDU。L36. After the RLC on the user side submits the verified SDU to the PDCP layer, it notifies the PDCP layer that data forwarding is over, and submits the subsequently received SDUs in sequence.

L37、PDCP层根据收到的SDU的PDCP编号进行重排序。L37. The PDCP layer performs reordering according to the PDCP numbers of the received SDUs.

L38与L18相同。L38 is the same as L18.

方法实例四：上述方式13、方式21和方式33的组合。Method example 4: a combination of the above-mentioned method 13, method 21 and method 33.

L41和L42与L11和L12相同。L41 and L42 are the same as L11 and L12.

L43、目标eNodeB在向用户侧发送SDU的过程中，在发出最后一个需要转发的SDU的各个PDU之后，在从aGW传输来的第一个SDU的最后一个PDU的包头中添加指示比特，或者通过显示信令方式标识从aGW传输来的第一个SDU的最后一个PDU，以告知用户侧转发数据的结束点。L43. In the process of sending SDUs to the user side, the target eNodeB, after sending each PDU of the last SDU that needs to be forwarded, adds an indication bit to the header of the last PDU of the first SDU transmitted from the aGW, or through The explicit signaling mode identifies the last PDU of the first SDU transmitted from the aGW, so as to inform the user side of the end point of the forwarded data.

L44和L45与L14和L15相同。L44 and L45 are the same as L14 and L15.

L46、用户侧RLC将RLC编号小于被标识的PDU的RLC编号对应的SDU都递交到PDCP层或结束RLC排序时，通知PDCP层乱序接收结束，并按序递交后续收到的SDU。L46. When the RLC on the user side submits the SDUs corresponding to the RLC numbers whose RLC number is smaller than the identified PDU to the PDCP layer or ends the RLC sorting, it notifies the PDCP layer that the out-of-order reception is completed, and submits the subsequent received SDUs in order.

L47、PDCP层根据收到的SDU的PDCP编号进行重排序。L47. The PDCP layer performs reordering according to the PDCP numbers of the received SDUs.

L48、PDCP层收到用户侧RLC实体的通知后，将排序后的SDU送到更高层，挂起其排序功能。L48. After receiving the notification from the user-side RLC entity, the PDCP layer sends the sorted SDUs to a higher layer and suspends its sorting function.

方法实例五：上述方式14、方式21和方式33的组合。Method Example 5: Combination of the above method 14, method 21 and method 33.

L51和L52与L11和L12相同。L51 and L52 are the same as L11 and L12.

L53、目标eNodeB在向用户侧发送SDU的过程中，在发出最后一个需要转发的SDU的各个PDU之后，在从aGW传输来的每个SDU的最后一个PDU的包头中添加指示比特；或者，通过显示信令方式标识从aGW传输来的每个SDU的最后一个PDU。当目标eNodeB收到任一一个被标识的PDU被正确接收的响应消息后，停止添加指示的操作。L53. In the process of sending the SDU to the user side, the target eNodeB, after sending each PDU of the last SDU that needs to be forwarded, adds an indication bit to the header of the last PDU of each SDU transmitted from the aGW; or, by The explicit signaling method identifies the last PDU of each SDU transmitted from the aGW. When the target eNodeB receives a response message indicating that any one of the identified PDUs has been received correctly, it stops the operation of adding indications.

L54和L55与L14和L15相同。L54 and L55 are the same as L14 and L15.

L56、用户侧RLC将RLC编号小于被标识的PDU的RLC编号对应的SDU都递交到PDCP层或结束RLC排序时，通知PDCP层乱序接收结束，并按序递交后续收到的SDU；以及在收到被标识的PDU之后，向目标eNodeB返回正确接收的响应消息。L56. When the RLC on the user side submits the SDUs corresponding to the RLC numbers whose RLC number is smaller than the identified PDU to the PDCP layer or ends the RLC sorting, it notifies the PDCP layer that the out-of-order reception is completed, and submits the subsequently received SDUs in order; and After receiving the identified PDU, return a correctly received response message to the target eNodeB.

L57和L58与L47和L48相同。L57 and L58 are the same as L47 and L48.

方法实例六：上述方式13、方式24和方式34的组合。Method Example 6: Combination of the above method 13, method 24 and method 34.

L61至L63与L41至L43相同。L61 to L63 are the same as L41 to L43.

L64、用户侧RLC缓存目标eNodeB发来的PDU，并按照PDU的RLC编号排序，当用户侧RLC收到被标识的PDU后，对缓存中RLC编号小于被标识的PDU的RLC编号的PDU进行SDU校验，之后将RLC编号小于被标识的PDU的RLC编号对应的SDU一并向PDCP层递交。L64. The user-side RLC caches the PDUs sent by the target eNodeB, and sorts them according to the RLC numbers of the PDUs. When the user-side RLC receives the identified PDUs, it performs SDUs on the PDUs whose RLC numbers in the cache are smaller than the RLC numbers of the identified PDUs. Check, and then submit the SDU corresponding to the RLC number whose RLC number is smaller than the identified PDU to the PDCP layer.

L65、用户侧RLC将校验后的SDU递交给PDCP层之后，通知PDCP层转发数据结束，并按序递交后续收到的SDU。L65. After the RLC on the user side submits the verified SDU to the PDCP layer, it notifies the PDCP layer that data forwarding is over, and then submits the subsequently received SDUs in sequence.

L67和L68与L37和L38相同。L67 and L68 are the same as L37 and L38.

方法实例七：上述方式11、方式21和方式31的组合；并且PDCP层记录并根据映射关系校验乱序SDU。Method example 7: a combination of the above method 11, method 21 and method 31; and the PDCP layer records and checks the out-of-sequence SDUs according to the mapping relationship.

L71至L74与L11至L14相同。L71 to L74 are the same as L11 to L14.

L75与L15相同；并且PDCP层记录收到的PDU的PDCP编号与RLC编号的映射关系。L75 is the same as L15; and the PDCP layer records the mapping relationship between the PDCP number of the received PDU and the RLC number.

L76与L16相同。L76 is the same as L16.

L77、PDCP层先根据该映射关系及转发数据结束点校验在转发数据结束点之前收到的PDU对应的SDU是否为乱序SDU。之后，PDCP层再根据收到的PDU的PDCP编号进行重排序。L77. The PDCP layer first checks whether the SDU corresponding to the PDU received before the end point of the forwarded data is an out-of-sequence SDU according to the mapping relationship and the end point of the forwarded data. After that, the PDCP layer performs reordering according to the PDCP numbers of the received PDUs.

L78与L18相同。L78 is the same as L18.

方法实例八：上述方式11、方式21和方式31的组合；并且用户侧RLC记录收到完整的并已经递交PDCP层的PDU的RLC编号，以及校验乱序SDU。Method Example 8: a combination of the above-mentioned method 11, method 21 and method 31; and the RLC at the user side records the RLC number of the received complete PDU that has been submitted to the PDCP layer, and checks the out-of-sequence SDU.

L81至L874与L11至L14相同。L81 to L874 are the same as L11 to L14.

L85与L15相同；并且用户侧RLC记录收到完整的并已经递交PDCP层的PDU的RLC编号。L85 is the same as L15; and the RLC on the user side records the RLC number of the PDU that has received the complete PDU and has been submitted to the PDCP layer.

L86、与L16相同；并且在用户侧RLC实体正确接收目标eNodeB发来的被标识的PDU之后，根据所述记录的PDU的RLC编号及转发数据结束点校验在转发数据结束点之前收到的PDU对应的SDU是否为乱序SDU，如果有编号大于被标识的PDU的RLC编号N对应的SDU已经被重组递交到PDCP层，则将该SDU当作已经成功接收，如果有需要，则进一步更新相应定时器或窗口边界，同时告知PDCP层该SDU不是乱序SDU。L86, the same as L16; and after the RLC entity on the user side correctly receives the identified PDU sent by the target eNodeB, check the PDU received before the end point of the forwarded data according to the RLC number of the recorded PDU and the end point of the forwarded data Whether the SDU corresponding to the PDU is an out-of-sequence SDU. If the SDU corresponding to the RLC number N with a number greater than the identified PDU has been reassembled and delivered to the PDCP layer, the SDU is regarded as successfully received, and further updated if necessary Corresponding timer or window boundary, and at the same time inform the PDCP layer that the SDU is not an out-of-sequence SDU.

L87、PDCP层根据收到的SDU的PDCP编号进行重排序。L87. The PDCP layer performs reordering according to the PDCP numbers of the received SDUs.

L78与L18相同。L78 is the same as L18.

为了支撑本发明方法，本发明还提供了一种切换过程中转发数据的系统，参见图6所示，其包括排序单元，以及与所述排序单元相连的递交单元和转发数据结束点通知单元；进一步还包括与所述递交单元相连的第一触发单元，以及与所述排序单元相连的第二触发单元。In order to support the method of the present invention, the present invention also provides a system for forwarding data during switching, as shown in FIG. 6 , which includes a sorting unit, a delivery unit connected to the sorting unit, and a forwarding data end point notification unit; It further includes a first triggering unit connected to the submitting unit, and a second triggering unit connected to the sorting unit.

所述第一触发单元，其位于用户侧RLC层，用于在切换开始后收到目标eNodeB发来的第一个SDU时，触发所述递交单元。The first triggering unit is located at the user side RLC layer, and is configured to trigger the handover unit when the first SDU sent by the target eNodeB is received after the start of the handover.

所述递交单元，其位于用户侧RLC层，用于向PDCP层递交SDU。The submitting unit is located at the RLC layer on the user side and is configured to submit the SDU to the PDCP layer.

所述第二触发单元，其位于PDCP层，用于在用户侧进行RLC重置时，触发排序单元开始排序。The second triggering unit is located at the PDCP layer, and is configured to trigger the sorting unit to start sorting when RLC reset is performed on the user side.

所述排序单元，其位于PDCP层，用于以PDCP序号对转发数据结束点之前递交单元发来的SDU进行排序。The sorting unit is located at the PDCP layer, and is used to sort the SDUs sent by the submitting unit before the end point of forwarding data by PDCP sequence number.

所述转发数据结束点通知单元，其位于源eNodeB、目标eNodeB及用户侧RLC中，用于通知转发数据结束点。The forwarding data end point notification unit is located in the source eNodeB, the target eNodeB and the user-side RLC, and is used for notifying the forwarding data end point.

显然，本领域的技术人员可以对本发明进行各种改动和变型而不脱离本发明的精神和范围。这样，倘若本发明的这些修改和变型属于本发明权利要求及其等同技术的范围之内，则本发明也意图包含这些改动和变型在内。Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalent technologies, the present invention also intends to include these modifications and variations.