W 2

MIMO AIR INTERFACE UTILIZING DIRTY PAPER CODING

FIELD OF INVENTION

[0001] The present invention relates to wireless communication systems.

More particularly, the present invention relates to utilizing dirty paper coding (DPC) to increase data throughput in wireless communications.

BACKGROUND

[0002] Modern communication systems support deployments wherein single-source terminals are configured to communicate various types of information to multiple terminals. One example of such a deployment includes a cellular communication system, such as a universal mobile telecommunication system (UMTS), wherein a base station communicates with a plurality of user terminals. Another example deployment includes an access point transmitting to multiple terminals in a wireless local area network (WLAN) system. These single-source-to-multiple-terminal types of deployments are commonly referred to as "broadcasting", "multicasting", or more generally as "point-to-multipoint" (PtM) communications.

[0003] Traditional PtM communication systems may be classified into one of two categories. The first category includes deployments in which transmissions to various receivers are permitted to interfere with each other. An example of this first type of PtM deployment includes a traditional code division multiple-access (CDMA) system, wherein pseudo-random codes are used to code various communications prior to their transmission. Upon receiving the coded transmissions, the user terminals synchronize the codes of the various transmissions to recover the communicated data. Although pseudo-random coding in the transmission source is useful in mitigating cross-interference, the true burden of mitigating this interference lies in the receiving terminals. [0004] In the second, more frequently utilized category of PtM deployments, interference is altogether avoided via the use of orthogonal

transmissions. Examples of such, deployments include those specified by the Global System for Mobile Communications (GSM) standards, such as frequency division multiple access (FDMA) systems and time division multiple access (TDMA) systems. In these types of PtM communication systems, a transmitting terminal transmits various communication signals using mathematically orthogonal or non-interfering "spaces". These spaces may be defined as a frequency range (as in a FDMA system), a time-slot (as in a TDMA system), or a Walsh code (as in an orthogonal code CDMA system). Unlike the first category of PtM communication systems, the transmitting terminal in this second category is solely responsible for preventing signal interference. As a result, receiving terminals in such systems are typically no more complex than those used in basic point-to-point communication systems. Although this category of PtM systems is superior, it should be understood that the performance of such systems is limited by the number of available orthogonal spaces and/or dimensions. [0005] Referring now to Figure 1, a graph 100 illustrating achievable data transmission rates 101-105 to the two receivers, Rx A and Rx B, in a PtM system is shown. It should be understood that Figure 1 is for illustrative purposes only and that is does not represent actual test results.

[0006] If all the available transmission bandwidth in a Figure 1 system were allocated to say, receiver Rx A, Rx A would receive service at a highest achievable data rate Ci and Rx B would receive no data. Similarly, if all the available transmission bandwidth were allocated to receiver Rx B, Rx B would receive service at a highest achievable data rate C2 and Rx A would receive no data.

[0007] If receivers Rx A and Rx B were operating in, for example, a TDMA system (which is equivalent to time-sharing), they would be capable of achieving data transmission rates at and to the left of the solid line 101 on the graph 100. As time sharing represents a special case of orthogonal multiplexing, similar rates are achievable by any system which maintains orthogonality between transmission signals.

[0008] In a PtM system, where orthogonality between transmission signals is not maintained, the transmission performance to any number of receivers can suffer as compared to that of an orthogonal system, such as TDMA. To illustrate, reference is again made to Figure 1. Line 102, for example, may be representative of achievable data rates in a typical random code CDMA system utilizing a standard RAKE receiver. Line 103 may be representative of achievable data rates in a typical random code CDMA system utilizing a more advanced MMSE multi-user detector. As indicated by the graph 100, the achievable data rates represented by line 103 are superior to those represented by line 102. Neither provides, however, the performance of line 101, which as described above, represents an achievable performance rate of transmission signals that are maintained orthogonal to each other.

[0009] It is well known from information theory that data rates superior to that of orthogonal coding (e.g., TDMA) are achievable in PtM systems. These superior data rates may be represented, for example, by lines 104 and 105 of the graph 100 shown in Figure 1. To achieve these superior data rates 104, 105, however, requires the use of receivers having receiver structures that are far more advanced then those used in typical receivers. To illustrate, information- theoretic successive interference cancellation (IT-SIC) can improve the performance of a CDMA system to where it actually performs better than TDMA systems. While such a result is counterintuitive at first, it is noted that the performance of a TDMA system is limited by the availability of orthogonal or non-interfering spaces or time-slots. IT-SIC structures allow interference, but in a controlled manner, and shifts interference cancellation to the receivers. Utilizing such structures enables a CDMA system to achieve data rates beyond those achievable with TDMA systems, such as those represented by the line 104 on the graph in Figure 1.

[0010] There are several problems with this IT-SIC approach. First, it requires highly complex receivers. Providing such complex receivers is problematic particularly in modern cellular systems, wherein the receivers must

be able to be embedded in relatively small, inexpensive terminal units with limited battery life. In addition, these complex receivers must possess information regarding both their own communication channel and the communication channels of all other users in the system. Dissemination of such channel information in practical communication systems is highly challenging. [0011] The problems cited above may be addressed using a technique called dirty paper coding (DPC). It is known theoretically that DPC performs at least as well as IT-SIC and in many cases, optimizes data throughput as illustrated by dashed line 105 on the graph 100 in Figure 1. In addition to providing for superior system performance, DPC has the added benefit of being a transmit-side ("pre-coding") technique. In other words, as in traditional TDMA and FDMA systems, the burden and complexity of interference cancellation is dealt with in the transmitting terminal, but without the limitations of orthogonal spaces. As a result, receivers are only required to possess detailed information pertaining to their particular communications. Furthermore, because each receiver operates optimally without possessing details of transmissions intended for other receivers, DPC provides a methodology for hiding transmissions from unintended receivers, thus making it suitable to support data hiding, watermarking, and other security applications.

[0012] The term "precoding", as used herein, refers to the mutual coding of multiple data streams while in the transmitter in order to pre-cancel any interference the data streams may cause each other; as opposed to attempting to cancel interference at individual receiving terminals post-transmission. It should be understood that pre-coding does not specifically imply that further coding steps will be performed, although further coding functions are possible. [0013] While recent analysis of DPC has yielded significant progress in the theoretical understanding of this technique, little is understood about how to build practical communication systems with DPC.

[0014] Accordingly, it is desirable to have a method and apparatus that utilizes DPC to optimize system performance and improve the signal quality of

transmission signals. It is further desirable to have a method and apparatus that utilizes DPC to support security-oriented applications such as data-hiding and watermarking.

[0015] It is well known that the use of multiple antennas provides for increased data throughput (via multiplexing) and improved signal quality of transmission signals (via diversity and/or beamforming). In addition, multiple antenna implementations, such as those found in MIMO (Multiple-Input Multiple-Output) and/or Smart antenna systems, enable transmission signals to be separated in a spatial domain (e.g., Spatial Division Multiple Access). To exploit these multiple antenna advantages, a MIMO technique, for example, may be combined with a signal processing technique such as FEC (forward error correction) coding and/or a radio air interface such as CDMA. There does not exist, however, a means for further improving multiple antenna systems utilizing DPC.

[0016] Accordingly, it is further desirable to have a method and apparatus that combines DPC with multiple antenna implementation(s) to improve both system performance and security enhancements.

SUMMARY

[0017] The present invention relates to a method and apparatus for optimizing system performance utilizing dirty-paper-coding (DPC) and/or multiple antennas to support security-oriented applications such as data-hiding and watermarking. A transceiver comprises a data processing module, a security processing module, a medium access control (MAC) module, a DPC module, and a smart antenna processor. The data processing module provides user data streams to the MAC module and channel state information to the smart antenna processor. The security processing module generates security data in the form of tokens and keys and provides the security data to the MAC module. In addition, the security module provides security policy data to the smart antenna processor. The MAC module determines data transmission rates for the user data streams

and the security data. In addition, the MAC module allocates the data streams and security data to transmission channels for transmission. The DPC module encodes the security data onto the user data streams and provides the encoded data streams to the smart antenna processor. The smart antenna processor generates pre-coding coefficients for use by the DPC module and transmits the security encoded data streams.

BRIEF DESCRIPTION OF THE DRAWINGS

[0018] A more detailed understanding of the invention may be had from the following description of a preferred embodiment, given by way of example and to be understood in conjunction with the accompanying drawings wherein: [0019] Figure 1 is a graph illustratively depicting achievable data rate regions to the two receivers under different precoding schemes; [0020] Figure 2 is a block diagram presenting the four major components of a dirty-paper coding (DPC) enabled transceiver;

[0021] Figure 3 illustrates a DPC-enabled transceiver which utilizes a plurality of DPC coders to encode data onto beams for transmission; and [0022] Figure 4 illustrates a DPC-enabled transceiver which utilizes a single QR-based DPC precoder to encode data onto beams for transmission.

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS [0023] Hereafter, the terminology "wireless transmit/receive unit" (WTRU) includes but is not limited to a user equipment, a mobile station, a fixed or mobile subscriber unit, a pager, or any other type of device capable of operating in a wireless environment. When referred to hereafter, the terminology "base station" includes but is not limited to a Node-B, a site controller, an access point or any other type of interfacing device in a wireless environment. [0024] Referring now to Figure 2, a block diagram 200 presenting four major components of a DPC-enabled transceiver is shown. These four (4) components include: 1) a DPC module 205; 2) a data processing module 201; 3) a

security processing module 203; and 4) a smart antenna processor 207. Included in the DPC module 205 is a low level medium access control (MAC) sub module 205a and a DPC component 205b. The security processing module 203 includes a token generation sub-module 203a. In addition to the four main components 201, 203, 205, 207 depicted in Figure 2, a DPC-enabled transceiver may further comprise a plurality of other components and sub-modules (not shown), depending on the particular requirements of such a device. [0025] The Data processing module 201 provides user data streams to the

MAC sub-module 205a. Also received at the MAC sub-module 205a are security tokens and keys (for each user in the PtM system) generated by the token generation sub-module 203a. Upon receiving this data, the low level MAC sub- module 205a determines a data transmission rate for both the user data and the tokens/keys. The MAC sub-module 205a then allocates the security data and user data streams to particular channels or beams for transmission. It is noted that the DPC module in Figure 2 and in the subsequent Figures does not need to distinguish between data which originated in a data processor and data which originated in a security module. The same DPC processing is applied to both types of information streams and the nature of the DPC processing is completely dependent upon the destination of the data, the pre-coding order, data rates and quality of service (QoS) targets, as determined by a MAC module. [0026] In the block diagram 200, the DPC component 205b of the DPC module 205, utilizes pre-coding coefficients from the smart antenna processor 207, and encodes or watermarks the security data 77a onto the user data streams using a DPC algorithm. These token-embedded data streams are then provided to the smart antenna processor 207 for further processing prior to transmission. [0027] In the smart antenna processor 207, high level security policy data and user channel state and location information are received from the security and data processing modules 203, 201, respectively. Utilizing this information, the smart antenna processor 207 provides beam-steering, pre-equalization, eigen-

beam-steering and other smart antenna functions for the transmission of the token embedded user data streams.

[0028] Referring now to Figure 3, a DPC-enabled transceiver 300 incorporating the four major components described with reference to Figure 2 is shown. As configured, the transceiver 300 performs DPC processing on a per- transmission beam basis. In addition to the four major components described above, the present transceiver 300 further comprises a plurality of sub-modules. The data and security processing modules 301, 303 of the present transceiver 300 correspond to similar modules 201, 203 of Figure 2. Included in the security processing module 303 is a token and key generation sub-module 303a for generating security tokens and keys on a per-user basis.

[0029] The MAC module 305 and DPC beam coders 307i..307 _n of the present transceiver 300 correspond to the MAC module 205 of Figure 2, but are depicted as separate modules. As previously discussed, these modules may comprise a single component, as depicted in Figure 2, or multiple components, as in the present transceiver 300. It should be noted that DPC coding functions of the present transceiver 300 are performed by a plurality of DPC coders 307i..307n, one per transmission-beam, rather than by a single DPC component. [0030] Also included in the present transceiver 300 are a plurality of pre- distortion antenna modules 309i..309 _n , transmit and receive antennas 311, 313, and a receive processing module 321, collectively considered the smart antenna component of the transceiver 300. The remaining modules, i.e., the channel estimation module 325, the pre-distortion filter computation module 327, the eigen-decompose module 331, and the beam identification module 333 are additional components of the transceiver 300, which may collectively comprise part of the smart antenna component.

[0031] User data streams from the data processing module 301 and security tokens and keys from the token generation sub-module 303a are provided to the MAC module 305. The MAC module 305 calculates appropriate data rates and allocates the security and user data to transmission-beams for

transmission. These transmission beams are identified and provided to the MAC module 305 by the beam identification module 333. The user and security data, data rates and beam allocation information is then output to the plurality of DPC coders 307i..307 _n on a per-beam basis. That is, all of the information that will be transmitted on a particular beam is output to a particular DPC coder 307i..307 _n for encoding. It should be noted that a user's data may be allocated to multiple beams for transmission in order to achieve multiplexing gains. Accordingly, if a user's data is allocated to multiple beams, such data will be DPC processed by a corresponding number DPC coders.

[0032] Upon receiving the MAC module 305 output, each DPC coder

307i..307 _n encodes the provided security data onto user data to form token- embedded data streams which will be transmitted using a previously assigned transmission-beam. As previously described, each DPC coder 307i..307 _n only processes data for one particular transmission beam. The token-embedded data streams are then output from the DPC coders 307i..307 _n to the plurality of pre- distortion antenna modules 309i..309 _n for further processing prior to transmission. Each pre-distortion antenna module 309i..309 _n receives input from a single DPC coder 307i..307 _n .

[0033] Also received in the pre-distortion antenna modules 309i..309 _n are pre-distortion filter coefficients from the pre-distortion filter computation module 327. The pre-distortion antenna modules 309i..309 _n apply (or multiply) these filter coefficients to the token-embedded data streams output from the DPC coders 307i..307 _n to yield scaled versions of such data streams. In particular, these filter coefficients may be utilized as zero-forcing and MMSE linear equalizers across space, time, and frequency domains. Since the filter coefficients are preferably complex values, the scaled output of the pre-distortion antenna modules 309i..309 _n will likewise preferably be complex values. This scaled or filtered output is then sent from the pre-distortion antenna modules 309i..309 _n to the transmit antennas 311, 313 for wireless transmission. It should be noted that the pre-distortion filter computation module 327 may also provide ISI

information to the pre-distort antenna modules 309i..309 _n , if the antenna modules 309i..309 _n are so configured to incorporate ISI pre-distortion. [0034] Received in the transmit/receive antennas 311, 313 are pilot signals transmitted from beam-receiving receivers (not shown). These pilot signals are processed by the receiver processing module 321 and input into the channel estimation module 325 where they are utilized for channel estimation. [0035] The channel estimation module 325 outputs channel estimates to the pre-distortion filter computation module 327, where such estimates are utilized to derive filter coefficients. As previously described, these filter coefficients are output to the pre-distortion antenna modules 309i..309 _n , where they are applied separately to each of the token-embedded data streams. The pre-distortion filter computation module 327 also provides its output to the eigen- decompose module 331.

[0036] In the eigen-decompose module 331, eigen-decomposition of a

MIMO channel is performed according to a H = USV ¹¹ decomposition, wherein H is the channel matrix of the MIMO channel, S is a diagonal matrix and U and V are unitary matrices of left and right eigen- vectors (eigenmodes) of the MIMO channel. This decomposition effectively decomposes the "intertwined" MIMO channel as observed by receive antennas 311, 313 into effective non-interfering or orthogonal (or nearly-orthogonal) sub-channels whose channel gains are given by the entries of S. The eigen-vectors provide the coefficients that are applied to the transmit/receive antennas 311, 313 in order to actually "form" orthogonal channels. Once eigen-composition is complete, the eigen-decompose module 331 provides its output to the beam identification module 333. [0037] The beam identification module 333 takes the orthogonal subchannels that are output from the eigen-decomposition module 331 and combines them into beams onto which user data may be transmitted. Each formed beam may consist of one or several of these sub-channels. In this regard, beamforming is used to separate between groups of users. For beams that are not fully orthogonal, i.e., nearly-orthogonal beams, any inter-beam interference is ignored.

As previously discussed, the MAC module 305 computes the amount of user data that each beam can support and subsequently, allocates data to the beams accordingly. It should be noted that a user's data may be allocated to more than one beam in order to achieve multiplexing gains. Any resulting inter-symbol interference may be resolved by channel pre-distortion, although partial ISI cancellation may be incorporated into the DPC encoders using, for example, Tomlinson-Harashima type precoding.

[0038] In the present embodiment, the MAC module 305 performs several complex operations including the allocation of users between transmission beams, wherein users whose spatial signatures are highly aligned are allocated to the same beam, while users whose spatial signatures are nearly orthogonal are allocated to appropriate, different beams. Since each user may have multiple eigenmodes, such eigenmodes may be in several different beams. As a result, a given user's data signal may similarly be allocated to different beams. Other complex operations performed by the MAC module 305 include deciding how to form the transmission beams and how the users should be allocated to such beams. For example, the MAC module 305 may decide that a single beam is appropriate, in which case the system transmitter behaves as a DPC-enabled transmitter with no beamforming. Alternatively, the MAC module 305 may determine that the number of required beams should be equal to the number of simultaneously active data streams, in which case no DPC is required and the system is equivalent to a beamforming enabled transmitter with no DPC. [0039] It is anticipated that there will be users whose spatial signatures are not sufficiently aligned with any of the transmission beams to be properly allocated to such beams. In such a scenario, the MAC module 305 develops several alternative beam combinations which are used in a time-shared fashion. Users are then separated according to which beam patterns their spatial signatures are compatible with and transmitted in the appropriate time-slots. [0040] In sum, the MAC 305 module of the present embodiment is preferably configured to: 1) identify all required user signatures; 2) develop

appropriate beam patterns and organize such beam patterns into compatible groups; 3) allocate users to appropriate beams; and 4) determine the rate allocated to each user while considering that DPC is used within each beam. [0041] Referring now to Figure 4, an alternate embodiment of a DPC- enabled transceiver 400 is shown. As configured, the transceiver 400 performs DPC precoder processing collectively on all transmission beams utilizing a single QR-based DPC precoder 407. Included in the present transceiver 400 are data and security processing modules 401, 403. The security processing module 403 includes a token and key generation sub-module 403a for generating security tokens and keys.

[0042] Also included in the present transceiver 400 are a plurality of pre- distortion antenna modules 409i..409 _n , transmit and receive antennas 411, 413, and a receiver processing module 481, collectively considered the smart antenna component of the transceiver 400. The remaining modules, i.e., the channel estimation module 485, the pre-distortion filter computation module 487, the eigen-decompose module 491, and the beam identification module 493 are addition components of the transceiver 400, which may collectively comprise part of the antenna component.

[0043] User data streams from the data processing module 401 and security tokens and keys from the token generation sub-module 403a are provided to the MAC module 405. The MAC module 405 calculates appropriate data rates and allocates the security and user data to transmission-beams for transmission. These transmission beams are identified and provided to the MAC module 405 by the beam identification module 493. The user and security data, data rates and beam allocation information is then output to the QR-based DPC precoder 407. This output from the MAC module 405 to the QR-based DPC precoder 405 includes a single eigenmode virtual data stream, which provides data for each user. Although data from a particular user may appear in multiple beams, since a single DPC precoder 407 is utilized, the data path is the same.

[0044] Unlike the MAC 305 of transceiver 300 previously described, the

MAC module 405 of the present embodiment does not need to allocate signal transmissions to different DPC coder modules. Instead, a QR-based approach (utilizing a single QR-based DPC pre-coder 407) is utilized, which is particularly useful for single antenna WTRUs and multi-antenna base stations. This QR- based approach is extended to multi-antenna WTRUs according to the following process. First, each multi-antenna WTRU's channel is decomposed to determine its eigenmodes and the singular values (i.e., channel powers) associated with these eigenmodes. Next, each eigenmode is treated as a channel to a "virtual" single-antenna WTRU. A QR-based decomposition approach is then applied to schedule transmissions to these virtual WTRUs. Each WTRU is "imagined" to be comprised of several virtual WTRUs, each equivalent to a single eigenmode. Accordingly, a controlling MAC controls the decomposition as in the transceiver 300 described with reference to Figure 3.

[0045] It should be noted that, in such a deployment, the number of simultaneous virtual single-antenna WTRUs to which a base station may transmit is limited by the number of transmit antennas at the base station. As a result, the MAC is required to time-share WTRU transmissions according to various parameters. Such parameters include the relative power of the channels (i.e., eigenmodes) to the virtual WTRUs and their cross-interference, which if optimized, maximizes the total data rate delivered in any given time slot. The MAC also applies a set of "fairness criteria" in scheduling transmissions which takes into account factors such as data rate and quality of service (QoS) requirements of the virtual WTRUs, previously used data rates for particular WTRUs, signal-to-noise ratios (SNR) of given eigenmodes, etc. [0046] Upon receiving the MAC module 405 output, the QR-based DPC precoder 407 encodes the provided security data onto the user data to form token- embedded data streams which will be transmitted using a previously assigned transmission-beam. As previously described, the QR-based DPC precoder 407 collectively processes data for all transmission beams. These token-embedded

W

data streams are then output from the DPC precoder 407 to the plurality of pre- distortion antenna modules 409i..409 _n , for further processing prior to transmission.

[0047] Also received in the pre-distortion antenna modules 409i..409 _n are pre-distortion filter coefficients from the pre-distortion filter computation module 487. The pre-distortion filter computation is performed jointly over all the antennas and beams. The pre-distortion antenna modules 409i..409 _n then multiply these calculated filter coefficients to the token-embedded data streams output from the QR-based DPC precoder 407 to yield scaled versions of such data streams. In particular, these filter coefficients may be utilized as zero-forcing and MMSE linear equalizers across space, time, and frequency domains. The QR decomposition of the channel matrix is utilized for dirty paper coding. Since the filter coefficients are preferably complex values, the scaled output of the pre- distortion antenna modules 409i..409 _n will likewise be complex values. This scaled or filtered output is then sent from the pre-distortion antenna modules 409i..409 _n to the transmit antennas 411, 413 for wireless transmission. It should be noted that the pre-distortion filter computation module 487 may also provide ISI information to the pre-distort antenna modules 409i..409 _n , if the antenna modules 409i..409 _n are so configured to incorporate ISI pre-distortion. [0048] Received in the transmit/receive antennas 411, 413 are pilot signals transmitted from beam-receiving receivers (not shown). These pilot signals are processed by the receive processing module 481 and input into the channel estimation module 425 where they are utilized for channel estimation. [0049] The channel estimation module 485 outputs channel estimates to the pre-distortion filter computation module 487, where such estimates are utilized to derive filter coefficients. As previously described, these filter coefficients are output to the pre-distortion antenna modules 409i..409 _n , where they are applied separately to each of the token-embedded data streams. The pre-distortion filter computation module 487 also provides its output to the eigen- decompose module 491.

[0050] In the eigen-decompose module 491, eigen-decomposition of a

MIMO channel is performed according to a H = USV ¹¹ decomposition, wherein H is the channel matrix of the MIMO channel, S is a diagonal matrix and U and V are unitary matrices of left and right eigen-vectors (eigenmodes) of the MIMO channel. This decomposition effectively decomposes the "intertwined" MIMO channel as observed by receive antennas 411, 413 into effective non-interfering or orthogonal (or nearly-orthogonal) sub-channels whose channel gains are given by the entries of S. The eigen-vectors provide the coefficients that are applied to the transmit/receive antennas 411, 413 in order to actually "form" these orthogonal channels. Once eigen-composition is complete, the eigen-decompose module 491 provides its output to the beam identification module 493. [0051] The beam identification module 493 takes the orthogonal subchannels that are output from the eigen-decomposition module 491 and combines them into beams onto which user data may be transmitted. Each formed beam may consist of one or several of these sub-channels. In this regard, beamforming is used to separate between groups of users. For beams that are not fully orthogonal, i.e., nearly-orthogonal beams, any inter-beam interference is ignored. As previously discussed, the MAC module 407 computes the amount of user data that each beam can support and subsequently, allocates data to the beams accordingly.

[0052] While the above description was presented in the context of a base station communicating with a plurality of WTRUs, it should be noted that the apparatus and techniques described herein are also generally applicable to single multi-antenna transmitters communicating a plurality of data streams to one or a plurality of receivers, each with one or several receive antennas. [0053] Also, the various DPC beamforming and pre-equalization enabled transceivers described herein support a variety of data services and security/watermarking services. Such transceivers use a MAC which supports viewing WTRUs that appear from a signal propagation point as being comprised of multiple virtual WTRUs, each of which is equivalent to a single eigenmode. In

this sense, the MAC operates on the virtual WTRUs. Further, the MAC is enabled to utilize a set of "fairness criteria" in scheduling transmissions to the virtual WTRUs.

[0054] The features of the present invention may be incorporated into an

IC or be configured in a circuit comprising a multitude of interconnecting components.

[0055] Although the features and elements of the present invention are described in the preferred embodiments in particular combinations, each feature or element can be used alone (without the other features and elements of the preferred embodiments) or in various combinations with or without other features and elements of the present invention.