Power-Efficient Intelligent Reception
FIELD OF THE INVENTION
The present invention relates to a method, system, apparatus, receiver module, and computer program product for providing power-efficient reception.
BACKGROUND OF THE INVENTION
Spectrum availability at frequencies that can be economically used for wireless communications is not satisfactory. This problem becomes aware, for example, when referring to the FCC (Federal Communications Commission) frequency chart which can be accessed, for example, by using the Internet address: http://www.fcc.gov/oet/info/database/spectrum/. The FCC frequency chart indicates multiple allocations over all available frequency bands. As a result, there is fierce competition for the use of spectra, especially in the bands below 3 GHz. However, according to D. Cabric, S. M. Mishra, R. W. Brodersen, "Implementation Issues inSpectrum Sensing for Cognitive Radios", 38th Annual Asilomar Conference on Signals, Systems and Computers, November 2004, actual measurements taken in an urban setting reveal a typical utilization of 0.5% in the 3-4 GHz frequency band. The utilization even drops to 0.3% in the 4-5 GHz band. Thus, it can be concluded that spectrum shortage is partially the result of regulatory and licensing processes.
The current approach for spectrum sharing is regulated so that wireless systems are assigned fixed spectrum allocations, operating frequencies and bandwidths, with constraints on power emission that may limit their range. Therefore, most communications systems are designed to achieve best possible spectrum efficiency within the assigned bandwidth using sophisticated modulation, coding, multiple antennas and other techniques. The most advanced systems are approaching Shannon's channel capacity limit, so further increase in capacity would require additional system bandwidth. On the other hand, the discrepancy between spectrum allocation and spectrum use suggests that spectrum shortage could be over- come by allowing more flexible usage of a spectrum. Flexibility would mean that radio terminals could find and adapt to any immediate local spectrum availability.
In FCC. Et docket no. 03-322, Notice of Proposed Rule Making and Order, De- cember 2003, a new radio class, so-called "cognitive radio", is described, that is able to reliably sense the spectral environment over a wide bandwidth, detect the presence/absence of legacy users (primary users) and use the spectrum only if the communication does not interfere with primary users.
In general, a cognitive radio - as its name implies - carries a level of cognition or intelligence that permits decision-making and learned patterns of behaviour. According to the Institute of Electrical and Electronic Engineers (IEEE), the congni- tive radio is a radio transmitter that is designed to intelligently detect whether a particular segment of the radio spectrum is currently in use and to jump into (or out of) a temporarily unused spectrum very rapidly without interfering with transmissions of other users. To achieve this, the wireless network or a wireless node itself are configured to change particular transmission parameters to execute tasks efficiently without interfering with licensed users. The parameter alteration can be based on observations of several factors, such as for example radio frequency spectrum, user behaviour, network state etc., so that the radio spectrum can be utilized more efficiently. More specifically, the radio transmitter (e.g., mobile terminal, mobile phone, user equipment, or the like) is configured to scan its environment, decide on the best frequency band as well as transmission standard and finally indicate to the other connection end (e.g., base station, access node, or the like) which transmit power, channel pre-equalization and pre-coding schemes should be used.
The cognitive radio concept requires flexible implementation on various layers. Especially the physical layer requires much more flexibility than today known from traditional non-cognitive radio standards. This flexibility can be achieved for the physical layer baseband processing by a software defined radio (SDR) implementation. SDRs rely on embedded software for their functionality and configuration. Assuming it is clear, which task a user wants to solve (voice call, data download, location tracking etc.), the cognitive radio needs to select a corresponding tech- nology (e.g., Global System for Mobile communication (GSM), Wireless Local Area Network (WLAN), Global Positioning System (GPS) etc.). In application specific integrated circuit (ASIC) implementations for conventional non-cognitive radios there is always the most critical case for wireless channel estimation plus channel decoding assumed, and thus maximum possible algorithm performance is targeted by implementing algorithms for worst case scenario, re- quiring highest complexity. As already mentioned above, in cognitive radios, a spectrum scanner identifies available spectrum resources and provides this information to a cognitive radio transmitter for corresponding transmission parameter selection.
Fig. 2 shows a diagram indicating processor load for different radio algorithms (decoding, channel estimation, frequency synchronization and timing synchronization) running concurrently on a floating point digital signal processor (DSP) of an orthogonal frequency division multiplexing (OFDM) SDR. It can be seen that channel estimation and decoding algorithms require most DSP processor load in this OFDM radio. The more critical the channel properties are the more sophisticated baseband algorithms need to be used for channel estimation and channel decoding. This leads to high processing loads and corresponding high power consumption, which is undesirable - especially for mobile terminals.
SUMMARY
It is therefore an object of the present invention to provide an enhanced method, device and/or system, by means of which power consumption can be reduced in intelligent receiver applications for wireless or wired communications.
This object is achieved by a method comprising:
• scanning at least one communication spectrum available for a desired communication;
• categorizing available communication channels based on at least one determined channel property;
• selecting a communication channel from said available communication channels in dependence on the result of said categorization; and • setting complexity of a receiver processing in dependence on the at least one channel property of said selected communication channel.
Additionally, the above object is achieved by an apparatus, comprising:
• a spectrum scanner configured to scan at least one communication spectrum available for a desired communication;
• a channel analyzer configured to categorize available communication chan- nels based on at least one determined channel property;
• a channel selector configured to select a communication channel from said available communication channels in dependence on the result of said categorization; and
• a setting unit configured to set a complexity of a receiver processing in dependence on the at least one channel property of said selected communication channel.
The apparatus may be configured as a receiver apparatus, a transceiver apparatus which comprises an additional transmitting functionality or unit, or as a receiver module provided as a part or integrated circuit of a more complex apparatus or system.
Moreover, the above object is achieved by a communication system comprising at least one of the above apparatus and at least one transmitter for communicating with said apparatus.
In addition, the above object is achieved by a computer program product compris- ing code means for producing the steps of the above methods when run on a computer device.
Accordingly, complexity of receiver processing can be reduced as much as possible based on the property of the selected channel, so that power consumption can be kept as low as possible. This leads to considerable power reduction in the processing for physical layer baseband, e.g., software processing in SDRs. Thus, if the receiver takes channel conditions or channel properties into account (e.g., by using channel state information or the like), it can choose a corresponding receiver algorithm complexity which would not be sufficient for worst case considerations (like in conventional (ASIC) receivers) but which fits to the actual channel properties to provide enough receiver performance. This choice of algorithms is inde- pendent of transmit scheme and can be realized in any type of intelligent receiver (such as cognitive radio or intelligent wired receivers), because such receivers can be configured to interpret channel conditions or properties and may employ flexible SDR implementation. These SDR or other software-based receiver implementions are flexible enough to configure receiver complexity during run-time.
The setting of the complexity may be performed independent from a transmission scheme selected for the desired communication.
Furthermore, the receiver processing may comprises at least one of channel estimation, channel decoding, filtering, and antenna selection. Then, complexity setting may be performed by at least one of setting complexity of said filtering by se- lecting a length of a digital filter chain, setting complexity of the channel estimation by selecting between zero forcing, linear interpolation and Wiener filtering, setting complexity of the channel decoding by selecting between direct decision processing, recurrent neural network processing, and Viterbi processing, and selecting a number of receiving antennas and/or a number of transmitting antennas. Of course, only exemplary setting examples are indicated above, which can be modified or replaced by other processing alternatives suitable to provide a range of different processing complexities (which define, e.g., amount of computational calculations, length of processed data words, amount of processing speed, etc.) for the receiving operations. Criteria for the setting operation may a minimum complexity sufficient to provide a predetermined receiver performance.
The above setting of the complexity may be achieved by supplying a predetermined set of processing parameters for said receiver processing. This set can then be used to select the processing algorithms or other criteria suitable for controlling complexity of the receiver processing.
In fast systems, such as software-base setting or control systems, the setting of the complexity may be performed during run-time of the desired communication. The above setting operations may be performed by the setting unit of the above apparatus.
The at least one determined channel property may be indicated by using a chan- nel state information.
Further advantageous modifications or developments are defined in the dependent claims.
BRIEF DESCRIPTION OF THE DRAWINGS
The present invention will now be described on the basis of embodiments with reference to the accompanying drawings in which:
Fig. 1 shows a schematic block diagram of a cognitive radio receiver apparatus according to an embodiment;
Fig. 2 shows a diagram indicating processing load of different receiver algorithms;
Fig. 3 shows a schematic diagram indicating selection of different receiver setups according to an embodiment;
Fig. 4 shows a schematic block diagram of a setting functionality for parameter- based selection of processing complexity according to an embodiment;
Fig. 5 shows a schematic flow diagram of a complexity setting operation according to an embodiment; and
Fig. 6 shows a schematic block diagram of a computer-implemented embodiment.
DESCRIPTION OF EMBODIMENTS
The embodiment will now be described based on a software-based cognitive radio apparatus, such as an SDR, which is configured to scan the environmental fre- quency spectrum and to decide on the best frequency band as well as transmission standard based on the scanning result. The radio apparatus may be any type of device, component, circuit, module etc., such as - but not limited to - a cellular telephone, a handheld computer, a multimedia device, or an integrated chip. The apparatus can be employed in any wired or wireless communication network which allows exchange of different types of data traffic based on a layered protocol stack which may be in conformity with the International Organization for Stan- dardization (ISO) Open Systems Interconnection (OSI) Reference Model (OSI/RM). Physical layer functions realize signaling for the specific transmission medium and are interfaced with a data link layer through a handshaking protocol. Even though cognitive radios are quite different from traditional wireless radios, the cognitive radio framework can be based on the ISO/OSI layering methodology.
Fig. 1 shows a schematic block diagram of an exemplary configuration of a cognitive radio transceiver apparatus according to an embodiment. The transceiver apparatus may be a mobile terminal, user equipment or other wireless transmit/receive unit which comprises a transmitter functionality or module and a re- ceiver functionality or module.
The radio apparatus comprises a wideband radio frequency (RF) front-end 10 capable of simultaneous sensing a wide spectrum, such as several GHz for example. This architecture is commonly proposed for SDRs. The wideband RF signal pre- sented at an antenna of such a front-end 100 includes signals from close and widely separated transmitters, and from transmitters operating at widely different power levels and channel bandwidths. The analog RF font-end 10 is connected via a analog-to-digital conversion portion (not shown) to a digital baseband processing unit 20 which could be implemented as a software-controlled processor or com- puter unit. In the baseband processing unit 20, converted reception signals are subjected to a receiver processing, while transmission signals to be transmitted by the RF front-end 10 are subjected to a transmitter processing. The receiver processing and transmitter processing may be performed in different processing paths or channels as indicated by the arrows in Fig. 1.
Furthermore, a channel analyzation unit 40 is provided, which controls a spectrum scanning operation performed by the RF font-end 10 and which analyses channel property or channel condition information CHI provided by the baseband processing unit 20 as a result of the spectrum scanning operation. The transmission and reception signals may be further processed by a digital signal processor 30 connected to the baseband processing unit 20. Cognitive radio communication depends on reliable detection of unoccupied spectrum. This requirement establishes a functionality on the physical layer for spectrum sensing over all available degrees of freedom (time, frequency, and space) in order to identify frequency bands currently available for transmission. A challenge of spectrum sensing is the detection of weak signals in noise with a very small probability of miss detection. Spectrum sensing requires the radio apparatus to receive a wideband signal through the RF front-end 10, sample it by high speed analog-todigital (A/D) converter (not shown), and perform measurements for detection of primary user signals. After identifying an available spectrum segment, the channel analyzation unit 40 controls the baseband processing unit 20 by supplying corresponding transmission processing parameters TxP to provide modulation schemes that provide best spectrum utilization and capacity while avoiding interference to any primary user. Furthermore, the desired transmission scheme should be flexible to allow assignments of any band to any user, and should be scalable with the number of users and bands. Thereby, a transmission signal can be created, which adaptively changes the occupied bandwidth without causing interference to any active primary users.
The wideband RF front-end 10 could be enhanced by an antenna array for spatial filtering. This array could be implemented as a phased antenna array where the antenna array coefficients are computed in the digital domain, e.g., by the baseband processing unit 20 in response to corresponding processing parameters supplied by the channel analyzation unit 40, and fed back to adjust the gains and phases of the antenna elements. A simple algorithm for computation of coefficients could be derived by noticing that strong primary users occupy distinct frequency bands and spatial directions of arrival. In order to obtain the estimate of angles of arrivals, the antenna array coefficients must sweep through many directions.
After reliable reception and sampling of a wideband signal, digital signal process- ing techniques can be utilized at the baseband processing unit 20 to further increase radio sensitivity by processing gain, and for primary user identification based on knowledge of the signal characteristics. Detection techniques can be based on matched filters, energy detectors, or cyclostationary feature detectors. Channel selection can be performed based on at least one of processing gain re- quired for a given probability of detection, sensitivity to unknown noise and interference, and implementation complexity. To keep the power consumption as low as possible, the channel analyzation unit 40 can be configured to decide on the best transmission parameters TxP (which may be used for controlling transmit power, channel coding etc.) and additionally could also choose the best quality wireless channel to ensure that the radio re- ceiver portion can reduce computational complexity as much as possible because of preferred wireless channel conditions.
As already decribed above in connection with Fig. 2, it can be seen that in this OFDM radio apparatus channel estimation and channel decoding algorithms require most processor load at the baseband processing unit 20. The more critical the channel properties are, the more sophisticated baseband algorithms need to be used for channel estimation and channel decoding. The more relaxed the wireless channel conditions are, the less mathematical complex algorithms are required. The channel analyzation unit 40 is therefore configured to first analyze channel conditions and then adapt the receiver algorithms to the channel condi- tions or properties of the selected channel. This can be achieved by applying suitable reception processing parameter RxP to the baseband processing unit 20.
Thus, the reception parameters RxP may be adapted to employ simple receiver algorithms, e.g., for channel estimation and channel decoding, if the cognitive ra- dio apparatus has been able to select a high quality channel (nearly ideal channel conditions). Otherwise, more complex algorithms are used in the receiver processing to handle more critical channel conditions, which have been identified before by the scanning operation of the channel analyzation unit 40.
The spectrum scanning operation controlled or performed by the channel analyzation unit 40 identifies available spectrum resources and provides information for corresponding selection of transmission parameter TxP (e.g., transmit power, channel coding scheme etc.). Additionally, independent of the transmitter parameter selection the channel conditions or properties (e.g. channel state information) is also into account for optimizing receiver processing at the baseband processing unit 20. Thus, the channel analyzation unit 40 utilizes the derived channel information for adapting transmitter and receiver operations.
Due to the fact that the channel information derived from the scanning operation is taken into account, receiver algorithm complexity can be adapted to the channel conditions or properties and not longer has to be defined based on worst case considerations. Thus, the reception processing parameters can be selected or set based on the actual channel properties to provide a predetermined receiver performance enough or sufficient for the desired purpose or application. The choice of receiver processing algorithms can thus be made independent from the transmit scheme and can be realized because the channel conditions are interpreted. This adaptive complexity of receiver processing can be employed in a flexible and fast manner, especially but not necessarily in software-based radio implementations. This provides sufficient flexibility to configure the receiver complexity during runtime.
The enhanced spectrum scanner operation performed or controlled by the channel analyzation unit 40 involves scanning the frequencies of the available spectrum and choosing best or desired frequency band and corresponding technology (e.g., GSM, 3G, WLAN etc.) depending on the desired application. Then, after deciding the desired frequency band, identifying the channel with the best channel properties inside that frequency band. Based on the channel conditions or properties (e.g. signaled channel state information), only a minimum of receiver algorithm complexity is selected, e.g., by applying a corresponding set of reception processing parameters RxP.
Fig. 3 shows a schematic diagram indicating an exemplary implementation example, in which different receiver setups are selected. After scanning the spectrum, the scanner function of the channel analyzation unit 40 identifies or detects the channel state information (CSI) of different free channels in one possible band. The better the channel conditions are, the less complex receiver algorithms are required. According to Fig. 3, the selected channels are categorized, e.g., based on the identified CSI, into "simple, nearly ideal channel", "increasing channel imperfections" (multipath, AWGN (Average White Gaussian Noise) type of noise, etc,), and "highly critical channel" (which requires high receiver complexity). Of course, other alternative or additional categories could be defined as well.
The left box shows a receiver setup with simplest low performance algorithms, which do not require much computational complexity and which are selected for the above case of "nearly ideal channel". The algorithms may involve simple filter processing with short digital filter chains (e.g., finite impulse response (FIR) filters), zero forcing processing as channel estimation, and direct decision processing as simple channel decoder. This is adequate because the channel is nearly ideal. Channel estimation is simple and channel decoding can be done via direct decision.
The central box of Fig. 3 belongs to a little more critical channel and is selected for the above case of "increasing channel imperfections", e.g., due to a time variant channel. The algorithms may involve improved filter processing with longer digital filter chains, linear interpolation processing as channel estimation, and parallel recurrent neural networks (RNN) as channel decoder. The amount of data processing in the receiver thus rises because channel estimation becomes more com- plex and channel decoding requires more knowledge about channel codes.
Finally, the right-hand box employs computationally complex algorithms and is selected for the above case of "highly critical channel", e.g., a fast time varying channel with strong interference. Only power hungry algorithms are able to equal- ize such a critical channel and to decode the code sequence correctly. The algorithms may involve steep filter processing with long digital filter chains, Wiener filter processing as channel estimation, and Viterbi processing as channel decoder. As an additional option, additional receiver branches (SIMO (Single Input Multiple Output) instead of SISO (Single Input Single Output) receiver) or at least one of a number receiving antennas and a number of transmitting antennas can be selected by providing a phased array antenna. In this regard it is noted that, besides the number of receiving antennas, also the number of transmission antennas has influence on the complexity of receiver processing. More specifically, the more transmitting antennas are combined, e.g., with good transmit diversity schemes like Alamouti, the more the receiver complexity could be reduced.
The choice which box (i.e., processing algorithms) should be chosen is done by the proposed cognitive spectrum scanning function of the channel analyzation unit 40, which does not select any free channel but a free channel with the best or predetermined properties, e.g., in terms of noise, multipath or Doppler influence.
The corresponding receiver algorithm complexity selection is independent of the transmitter specification, because all receiver algorithms fulfill the same receiver tasks (channel estimation, channel decoding, ...) with different mathematical com- plexity and different performance. Fig. 4 shows a schematic block diagram of a setting functionality for parameter- based selection of processing complexity according to an embodiment. The setting functionality can be provided in the channel analyzation unit 40 which may be adapted to generate based on a scanning input signal I, received from the RF font- end 10, a set of reception processing parameters RxP1 to RxPn used for controlling, establishing, or setting up the above or other processing algorithms at a cognitive receiver processing part 210 of the baseband processing unit 20. The cognitive receiver processing part 210 may be implemented as a software routine controlling a processing or computer unit of the baseband processing unit 20 based on the set of reception processing parameters RxP1 to RxPn, or as a hardware- implemented digital processing circuit which is controlled by the set of reception processing parameters RxP1 to RxPn.
Fig. 5 shows a schematic flow diagram of a complexity setting operation according to an embodiment. It may be provided as a software function or routine as a part of the overall spectrum scanner functionality implemented in the channel analyzation unit 40 and/or the baseband processing unit 20.
In step S101 , the analog RF front-end 10 is controlled to scan the radio spectrum. Then, in step S102, the baseband processing function 20 and/or the channel analyzation unit 40 make a rough channel analysis to categorize the actual or available channel(s) based on their conditions or properties, e.g., into quality levels as indicated in Fig. 3. In step S103, the channel analyzation unit 40 selects, after comparing all different available channels, the best, preferred or simplest channel with most ideal or at least sufficient properties. After selecting the channel, the channel analyzation unit 40 decides in step S104 on the required receiver processing complexity. This decision my be based on a fixed parameter allocation as for example indicated in Fig. 3. Thereafter, in step S105, the channel analyzation unit 40 informs transmitter and receiver processing portions of the baseband proc- essing unit 20 about the decided processing parameters, e.g., by setting the transmission and reception processing parameter TxP1 to TxPn and RxP1 to RxPn. Thus, based on the channel quality analysis, the receiver processing portion is controlled and thus gets an indication how complex the receiver algorithms should be.
Fig. 6 shows a schematic block diagram of a software-based implementation of the proposed functionalities for achieving channel-sensitive complexity adjustment. These functionalities can be implemented with a processing unit 210, which may be any processor or computer device with a control unit which performs control based on software routines of a control program stored in a memory 212. Program code instructions are fetched from the memory 212 and are loaded to the control unit of the processing unit 210 in order to perform the processing steps of the above functionalities described in connection with the respective blocks of Figs. 1 , 4, and 6. These processing steps may be performed on the basis of input data Dl and may generate output data DO, wherein the input data Dl may correspond to the input signal I, which may be derived from the spectrum scanning operation at the RF front-end 10, and the output data DO may correspond to the set of parameters RxP1 to RxPn and TxP1 to TxPn or any other control information provided to achieve complexity control.
To summarize, a method, system, apparatus, receiver module and computer pro- gram product have been described for providing a power-efficient reception, wherein a spectrum available for a desired communication is scanned and available communication channels are categorized based on at least one determined channel property. Then, a communication channel is selected from the available communication channels in dependence on the result of the categorization, and complexity of a receiver processing is set in dependence on the at least one channel property of the selected communication channel. Thus, complexity of the receiver processing is reduced as much as possible in the light of the property of the selected channel, so that power consumption can be kept as low as possible.
It is to be noted that the present invention is not restricted to the embodiments described above, but can be implemented in any communication apparatus with a receiver functionality for any type of wired or wireless application. As an example, at least one spectrum may be shared in a cable, optical fibre, or other type of electrical, magnetic, electro-magnetic or optical waveguide. The processing steps of Fig. 6 may be implemented as discrete digital circuits, modules or logical signal processing structures. The number of categories for channel selection and complexity allocation may differ from the example of Fig. 3 and may even be replaced by individual channel decisions made by the channel analyzation unit 40 based on a comparison of channel information, such as the CSI or other quality-based pa- rameters, derived from the spectrum scanning operation. The embodiment may thus vary within the scope of the attached claims.