AU2012200312B2 - Method and apparatus for providing efficient precoding feedback in a MIMO wireless communication system - Google Patents

Abstract

A multiple-input multiple-output (MIMO) scheme uses precoding and feedback in a wireless communication system including a transmitter and a receiver. The system may use either a single codeword (SCW) or a double codeword (DCW). The precoding scheme is based on transmit beamforming (TxBF). Combined differential and non-differential feedback with periodic resetting is considered. 7- CN CD. -cO 0 CD co CfoJ co co tc c coo _ pc Ln. c o mO

Description

Pool Seation 20 Regyistion 3.2(2) AUSTRALIA Patents Act 1990 ORIGINAL COMPLETE SPECIFICATION STANDARD PATENT Application Number: Lodged: Invention Title: Method and apparatus for providing efficient precording feedback in a MIMO wireless communication system The following statement is a full description of this invention, including the best method of performing it known to us: P11 1AHAU/0710 2 [0007] The efficient feedback can reduce feedback overhead or improve performance. A potential feedback overhead reduction is obtainable when using the Jacobi rotation for eigen-basis feedback. Additional overhead reduction is achievable using a differential feedback by an iterative approach for the Jacobi transform to track the delta of the eigen-basis and then provide feedback to the new 5 eigen-basis. [0008] It would be desirable to use differential feedback and iterative Jacobi rotation for potential feedback overhead reduction and performance improvement. Iterative Jacobi transform based feedback is a potential solution for a two or more transmit antenna MIMO proposal. [0008A] Reference to any prior art in the specification is not, and should not be taken as, an 10 acknowledgment or any form of suggestion that this prior art forms part of the common general knowledge in Australia or any other jurisdiction or that this prior art could reasonably be expected to be ascertained, understood and regarded as relevant by a person skilled in the art. SUMMARY [0009] The present invention evaluates performance of MIMO precoding scheme and consider the 15 effects of quantization, group feedback and feedback delay for MIMO precoding in wireless communication system including a transmitter and a receiver. The system may use either a single codeword configuration (SCW) or a double codeword (DCW) configuration. Singular value decomposition (SVD) can be used to generate the precoding matrix. The quantization for MIMO precoding or transmit eigen-beamforming (TxBF) can be codebook-based. Group feedback 20 considers one feedback per group of subcarriers or resource blocks (RBs). A codebook-based MIMO precoding scheme using combined differential and non-differential feedback is also provided. The precoding scheme may only use non-differential feedback. [0010] The present invention evaluates performance of MIMO precoding scheme and consider the effects of quantization, group feedback and feedback delay for MIMO precoding. SVD can be used 25 to generate the pre-coding matrix. The quantization for MIMO pre-coding or TxBF can be codebook-based. Group feedback considers one feedback per group of subcarriers or resource blocks (RB). We consider the codebook-based MIMO precoding scheme using combined differential and non-differential feedback.

3 The present invention provides a precoding feedback scheme based on Jacobi rotations for uplink MIMO. The present invention can also be applied to downlink MIMO where OFDM(A) is used. Combined differential and non-differential feedback with periodic resetting is considered. It is shown that the differential feedback with proper resetting improves performance. Differential 5 feedback requires considerably less, about 33%, feedback overhead than non-differential feedback while the performance is maintained. The performance degradation for MIMO precoding due to quantization, group feedback and feedback delay is studied. It is shown that the performance degradation due to quantization for MIMO precoding is within a fractional dB. The performance degradation of MIMO precoding due 10 to group feedback depends on the channel coherent bandwidth and the size of the feedback group. The loss is within 1 dB for feedback every 25 RBs. It is also shown that performance degradation due to feedback delay is within a fractional dB for low speed or shorter feedback delay such as 3 km/h or feedback delay of 2 transmission time intervals (TTIs). The performance degrades more as the speed or feedback delay increases. 15 In one aspect the present invention provides a method of providing precoding feedback including: receiving a plurality of feedback bits; updating a first precoding matrix based on the feedback bits, wherein the feedback bits are differential feedback bits; precoding a plurality of frequency domain data streams using the first precoding matrix; mapping the differential feedback bits to a delta precoding matrix by using a differential codebook; and generating a full precoding matrix based on 20 the delta precoding matrix. In a further aspect the present invention provides a receiver for providing feedback to a transmitter for updating a first precoding matrix used by the transmitter to precode a plurality of frequency domain data streams, the receiver including: a channel estimator configured to generate a channel estimate by performing a channel estimation on frequency domain data associated with a plurality 25 of time domain data streams transmitted by the transmitter; a feedback generator electrically coupled to the channel estimator, the feedback generator configured to generate feedback bits for transmission to the transmitter based on the channel estimate, wherein the feedback bits are differential feedback bits; the feedback generator including: a precoding matrix generator 3A configured to generate a delta precoding matrix based on the channel estimate; and a feedback bit generator electrically coupled to the delta matrix generator, the feedback bit generator being configured to generate and transmit feedback bits based on the delta precoding matrix. In another aspect the present invention provides a transmitter that performs precoding based on 5 feedback provided by a receiver, the feedback being generated based on a plurality of time domain data streams that the receiver receives from the transmitter, the transmitter including: a precoding matrix generator configured to receive feedback bits from the receiver and update a precoding matrix based on the feedback bits, wherein the feedback bits are either non-differential feedback bits or differential feedback bits; and a precoder electrically coupled to the precoding matrix generator, 10 the precoder being configured to precode a plurality of frequency domain data streams using the precoding matrix, the precoder including: a feedback bits to delta preceding mapping unit for mapping differential feedback bits to a delta precoding matrix; and a full precoding matrix generation and update unit for generating and updating a full precoding matrix based on the delta precoding matrix, wherein the precoder uses the full precoding matrix to precode the frequency 15 domain data streams. BRIEF DESCRIPTION OF THE DRAWINGS 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: 20 [0015] Figure 1 is a graph showing the frame error rate (FER) versus signal-to-noise ratio (SNR) using a Typical Urban 6 (TU-6) channel model. A comparison of ideal and quantized feedback is given; [0016] Figure 2 is a graph showing the frame error rate (FER) versus signal-to-noise ratio (SNR) using a Spatial Channel Model Extended C (SCME-C) channel model. A comparison of ideal and 25 quantized feedback is given. As observed there is less loss from quantized feedback for the SCME C channel model than the TU-6. channel model. This is due to correlation properties of the SCME C channel model; 3 The present invention provides a precoding feedback scheme based on Jacobi rotations for uplink MIMO. The present invention can also be applied to downlink MIMO where OFDM(A) is used. Combined differential and non differential feedback with periodic resetting is considered. It is shown that the 5 differential feedback with proper resetting improves performance. Differential feedback requires considerably less, about 33%, feedback overhead than non differential feedback while the performance is maintained. The performance degradation for MIMO preceding due to quantization, group feedback and feedback delay is studied, It is shown that the performance 10 degradation due to quantization for MIMO precoding is within a fractional dB. The performance degradation of MIMO precoding due to group feedback depends on the channel coherent bandwidth and the size of the feedback group. The loss is within 1 dB for feedback every 25 RBs. It is also shown that performance degradation due to feedback delay is within a fractional dB for low speed or 15 shorter feedback delay such as 3 km/h or feedback delay of 2 transmission time intervals (TTIs). The performance degrades more as the speed or feedback delay increases. In one aspect the present invention provides a method of providing precoding feedback including: 20 receiving a plurality of feedback bits; updating a first precoding matrix based on the feedback bits, wherein the feedback bits are either non-differential feedback bits or differential feedback bits; precoding a plurality of frequency domain data streams using the first precoding matrix; 25 receiving a plurality of time domain data streams, each time domain data stream including a cyclic prefix (CP); removing the CPs from the time domain data streams to generate a plurality of processed data streams; converting the processed data streams to frequency domain data; 30 performing channel estimation on the frequency domain data to generate a channel estimate; generating a second precoding matrix based on the channel estimate; and 3a generating and transmitting feedback bits based on the second preceding matrix. In a further aspect the present invention provides a receiver for providing feedback to a transmitter for updating a first precoding matrix used by the 5 transmitter to precede a plurality of frequency domain data streams, the receiver including: a channel estimator configured to generate a channel estimate by performing a channel estimation on frequency domain data associated with a plurality of time domain data streams transmitted by the transmitter; 10 a feedback generator electrically coupled to the channel estimator, the feedback generator configured to generate feedback bits for transmission to the transmitter based on the channel estimate, wherein the feedback bits are either non-differential feedback bits or differential feedback bits; a plurality of antennas configured to receive the time domain data streams; 15 a plurality of cyclic prefix (CP) removal units electrically coupled to respective ones of the antennas, each CP removal unit being configured to remove a CP from each of a plurality of time domain data streams received by the antennas to generate processed data streams; and a plurality of fast Fourier transform (FFT) units electrically coupled to 20 respective ones of the CP removal units and the channel estimator, each FFT unit being configured to convert the processed data streams to frequency domain data. In another aspect the present invention provides a transmitter that performs preceding based on feedback provided by a receiver, the feedback 25 being generated based on a plurality of time domain data streams that the receiver receives from the transmitter, the transmitter including: a precoding matrix generator configured to receive feedback bits from the receiver and update a preceding matrix based on the feedback bits, wherein the 30 feedback bits are either non-differential feedback bits or differential feedback bits; and 3b a precoder electrically coupled to the precoding matrix generator, the precoder being configured to precode a plurality of frequency domain data streams using the preceding matrix, the precoder including: a feedback bits to delta precoding mapping unit for mapping differential 5 feedback bits to a delta preceding matrix; and a full precoding matrix generation and update unit for generating and updating a full precoding matrix based on the delta precoding matrix, wherein the precoder uses the full preceding matrix to precode the frequency domain data streams. 10 BRIEF DESCRIPTION OF THE DRAWINGS A more detailed understanding of the invention may be had from the following description of a preferred embodiment, given byway of example and to be understood in conjunction with the accompanying drawings wherein: [0015] Figure 1 is a graph showing the frame error rate (FER) versus signal-to-noise 15 ratio (SNR) using a Typical Urban 6 (TU-6) channel model. A comparison of ideal and quantized feedback is given; Figure 2 is a graph showing the frame error rate (FER) versus signal-to noise ratio (SNR) using a Spatial Channel Model Extended C (SCME-C) channel model. A comparison of ideal and quantized feedback is given. As observed there 20 is less loss from quantized feedback for the SCME-C channel model than the TU 6. channel model. This is due to correlation properties of the SCME-C channel model; 25 [00171 Figure 3 is a graph comparing differential feedback and non differential feedback; (0018] Figure 4 is a graph of feedback using different resetting intervals; 100191 Figare 5 is a graph comparing difterential feedback with feedback delay for SOME-0 at a lower speed; {00201 Figure 6 is a graph of differential feedback and feedback delay for SOME-C at a high speed; and [0021] Figure 7 is a graph of non differential feedback and feedback delay for SOME-C at a high speed. [0022] Figure BA is a block diagram of a transmitter including a preceding matrix generator for processing differential or non-differential feedback bits in accordance with the present invention; [0023] Figures SB and SC show details of the precoding matrix generator of Figure 8A; [0024] Figure SA is a block diagram of a receiver including a feedback generator that generates the feedback bits processed by the preceding matrix generator of the transmitter of Figure BA in accordance with the present invention; [0025] Figures 9B and 90 show details of the feedback generator of the receiver of Figure 9A; [0026] Figures Z0A and 1OB show different embodiments of the preceding matrix generator used in the feedback generator of Figure 9B; [0027] Figures 10C and 1OID show different embodiments of the preceding matrix generator used in the feedback generator of Figure 90; [00281 Figure 11 shows a comparison of double codeword performance for single user MUMO (SU-MIMO) to single-input, multiple-output (SIMO) fbr the high data throughput SNR regions; and [00291 Figure 12 shows a comparison of the performance for single and double codewords using uplink preceding MIMO for two or more antennas at the WTRU and an evolved Node-B (eNodeBl) with an SOME-0 channel. -4t0080 DETAILED DESCRIPTION OF THE PREFERRED EMBOD1MNTS [0031] When referred to hereafter, the terminology "wireless trantsmit/receive unit (WTRU)" includes but is not united to a user equipment (UE), a mobile station, a fixed or mobile subscriber unit, a pager, a cellular telephone, a personal digital assistant (PDA), a computer, or any other type of user 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 (AP), or any other type of interfacing device capable of operating in a wireless environment. [0032] Non-differential feedback 10083] A Jacobi rotation is used to perform matrix diagonalization. The channel response matrix H (or the estimate of channel response matrix) can be decomposed into: v- = UDV", 1.quation (1) where U and V are unitary matrices, i.e., UU =-r and V"V = 1. D is a diagonal matrix that has singular values in the diagonal, V is the eigen-matrix (consisting of eigen-veotors) and can be used as a precoding matrix at the transmitter, and Vm is the Hormetian .of a preceding matrix (eigen-matrix) V. The channel correlation matrix R is defimed as: R = H 1 "H, Equation (2) which is the product of the Hermitian transpose of the channel response matrix H and the channel response matrix H- itself. The channel correlation matrix R can be decomposed into: R = VD 2 V". Equation(3) [00841 Jacobi rotation is used to perform the matrix diagonalization on the channel correlation matrix R such that: 12 - J"RJ. Equation(4) [0035] Diagonalization is a process of transforming any arbitrary matrix into a diagonal matrix. Diagonalization is typically used in wireless communications and signal processing applications to separate multiple signals -. 5and/or to separate the desired signal and interference. Equation (4) describes the process of diagonalizing the channel correlation matrix R into a diagonal matrix D . In Equation (4), the Jacobi rotation matrix J is multiplied with the channel correlation matrix R from the right-hand side, and the Hermitian transpose of Jacobi rotation matrix J is multiplied with the channel correlation matrix R from left-hand side. The resulting matrix is D' which is a diagonal matrix. When comparing Equations (1) and (3), it is observed that to diagonalize the channel response natrix H to find the eigen-matrix V is equivalent to diagonalize the channel correlation matrix R to End eigen-matrix V. Equation (8) can be rewritten as: V"RV =D* . Equation (5) [0036) When comparing equations (4) and (5), itis observed that the Jacobi matrix J becomes the eigen-matrix V when the channel correlation matrix R is diagonalized using eigen-value decomposition (or SVD) and Jacobi rotation for the diagonalization transform. The Jacobi rotation transform or preceding matrix (or the estimate of Jacobi rotation translonn or precoding matrix) for a 2x2 configuration is represented as: J( [ n Equation (6a) where 0 and 0 are estimates of parameters for the Jacobi rotation., The parameters d and 0 can be obtained by the equations 9 and 10. The parameters $ and 0 can also be obtained by solving the equation 6b below. V =L" iJ - J, )= L , d l Equation (6b) V21 V ' - sin(0) COo 1 [0037] The preceding matrix (eigen-matrix) V is represented as: V=[ 12 1 Equation(7) -6- 0038] The channel correlation matrix R is represented as: R- Equation(S) [0089] For non-differential feedback, the preceding matrix V feedback is performedL Since the preceding matrix V is equivalent to the Jacobi rotation matrix J by comparing Equations (4) and (5) as discussed in previous sections, the preceding matrix V can be transformed into the Jacobi rotation matrix J. Feeding back the preceding matrix V is equivalent to feeding back the Jacobi rotation matrix J or feeding back the parameters 0 and e of the Jacobi rotation matrix. The feedback of the preceding matrix V can be represented by two elements: 9 and /, instead of v11, v12, u21, and v22 (the elements or the eigen vectors of the preceding matrix V) or r11, r12, r21, and r22 (the elements of the channel correlation matrix R). The feedback of parameters of the inatrix transform (such as feedback of 0 and 9) is more efficient than the feedback of the entire preceding matrix, or the precoding vectors themselves (such as the feedback of the preceding matrix V or equivalently its elements v11, v12, v21, and v22, or the feedback of the channel correlation matrix R or equivalently its elements rl., r12, r2l, and r22), [0040] The Jacobi transform parameters $ and 0 can be computed using the following two equations;. tan(o) + ( -r)tan(O) 1 -0 ; and Equation(9) rld e =3 , Equation(10) where r, is the element of channel correlation matrix R that corresponds to the ih row and jtb column (00413 To further reduce feedback overhead, differential processing is introduced in which only the changes or differences of the parameters of matrix transform (AS and A ) between updates are computed, and fed back. -7- [00421 To. avoid error aceum-Lation and propagation introduced by differential processing, ai approach that combines differential and non differential feedback is considered in which a differential feedback with periodic error reset is proposed. [00431 Differential feedback [0044] The differential feedback using an iterative Jacobi transform is proposed. For feedback instance n, the Jacobi rotation J(n) is applied on channel correlation matrix R and is expressed byr J(n)" 1 R(n)J(n)= D. Equation(11) For the next feedback instance n+1, if the Jacobi rotation matrix is not updated, diagonalization of matrix R.2 using Jacobi rotation of feedback instance n can be expressed by: J(n)"R(n+1)/(ni. Equation (12) -P is not diagonal. However, when the channel changes slowly, bP is close to diagonal. When the channel is not changed, 5P is diagonal. When MIMO channels change, ZP is no longer diagonal. The precoding matrix and, therefore, the Jacobi rotation matrix, needs to be updated for correct diagonalization. Call A] (or AJ(n)) the differential precoding matrix (delta precoding matrix) that represents the delta of the feedback matrix update at feedback instance n. The parameters A 6 and Ag for Jacobi rotation transform of delta precoding matrix are sent back to the transmitter from the receiver. This is in contrast to the non differential feedback in which a full precoding matrix instead of the delta precoding matrix is fed back. The parameters 0 and 0 for Jacobi rotation transform of the full preceding matrix are fed back to the transmitter. When the channel changes, the Jacobi rotation or transform needs to be updated for correct diagonalization: AJ(n)"[Jn)"R(n +1)J(n)]AJ(n) = AJ(n)"IDAM(n) D Equation(13) -8where AJ(n) is the delta of the feedback update at feedback instance n. The differential feedback or delta feedback AJ(n) is estimated and computed at the receiver and is sent back to the transmitter from the receiver for updating the preceding matrix J(n) for the next preceding process J(n+1) at transmitter (and/or at the receiver if needed). {00451 The differential feedback or delta feedback A] can be obtained from Z9 where: 2 = d d - Equation(14) Idu d22 [0046] The following Equations (15) and (16) can be used to obtain the differential preceding matrix Al, (i.e., to obtain Ad and A 0- ): tan(Ay $ +("- {") tan(A)- 1= 0 ; and Equation (15) Alternatively, the differential feedback A] can be computed at the receiver by multiplying the Hernitian transpose of the previous preceding matrix J(n) with the preceding matrix J(n+X) by: AJ(n) J(n)" J(n +1), Equation(17) where J(n+1) can be computed from the correlation matrixR(n+1) at the receiver as described in Equations (2) and (4) for feedback instance n+L. The transmitter receives the feedback AJ(n) and uses it for the preceding matrix update for J(n+1). Note that the preceding matrix is denoted as J (which is equal to V as J and V are equivalent as discussed in previous sections). The previous preceding matrix J(n) at the transmitter is updated to obtain the next preceding inat-rix J(n+1). The transmitter first receives and decodes the feedback bits, and translates those feedback bits to a delta preceding matrix AJ. This can be performed at the transmitter by multiplying the previous preceding matrix J(n) that is used at the transmitter with the differential preceding matrix AJ(n) that is received; decoded and translated by the transmitter from the receiver by: J(n + 1) = J(n) -&(n) Equation (18) J(n+1) can be computed from R(.-+1), and R(n+l) is calculated from lH(n+L). (0047] Diagonalization is achieved using an updated differential precoding matrix AJ, as described by Equation (13), and the resulting equation can be rewritten as: J(n +1 R(n + I)J(N+ 1)= D, EquatiiO(19) where J(n-i-1) and V are related by Equation (18). [00481 Combined Differential and Non-Differential Feedback [0049] Note that both combined differential and noiidifferential feedback may be used with group feedback. Group feedback assumes that adjacent sub carriers or resource block (lB) wrn erbibit similar fading behavior and as such these techniques may be applied to them jointly. [0050] In general, differential feedback may be more suitable for low speed channels and non-differential feedback may be suitable for high speed channels. A combined differential and non-differential feedback may be considered for feedback overhead reduction and performance improvement. [00511 Differential feedback can be reset every N TTIXs, every N feedback intervals, every certain period of time or aperiodically for avoiding error accumulation or propagation due to differential prooessng. Nis a predetermined integer. At each reset, non-diXerential feedback is used. Non-differential feedback occurs every N TTIs or every N feedback intervals and differential feedback is used for the remaining TTs or feedback intervals. At the resetting period, the full preceding matrix is fed back while, between the resets or between non-differential feedback, only the delta preceding iatrix is fed back, [00521 The feedback overhead can be reduced. For differential feedback, less bits, (e.g., 2 bits), are required for quantization. For non-differential feedback, more hits, (e.g., 3 bits), are required for quantization. [0053] For example a codebook consisting of eight codewords which requires three (8) feedback bits for quantization is used for non-differential -10feedback, while four codewords are-usedfor differential feedback, which requires fewer feedbackbits (2 bits). The feedback can be based on averages over mrdtiple resource blocks (RBs), (e.g., 2, 5, 6, 10 RBs), where a RB is defined as a block with multiple subcarriers (e.g., 12 or 25 subcarriers). [0054 Two codebooks are used. The codebook, (differential codebook), used for quantization concentrates on the origin of the (0,0) plane for differential feedback, while the codebook, (non-diflerental codebook), for non-differential feedback is uniform with codewords evenly distributed. For one imlIementation, the differential codebook consists of four codewords. The non-differential codebook consists of eight codewords. A combined differential and non differential feedback can reduce the feedback overhead and improve the performance for the MIMO precoding. [00551 Simulation AssumRtions The simulation assumption and parameters used are given in Table 1 below. Parameter Assumption Carrier frequency 2.0 GHz Symbol rate 4.096 mnfiion symbols/sec Transmission bandwidth 5 MHz TTI lengh 0.5 is (2048 symbols) Number of data blocks per TTI 6 Number of data symbols per TTI 1536 Fast Fourier transform (FFT) block 512 size Number of copied subcarriers 256 Cyclic Prefix (CP) length 7.8125 psee (32 samples) Channel model Typcal Urban (TU6). SOM-C Antenna configurations 2 x 2 (MIMO) Fading correlation between p = 0 for TU6, and SCME-C transmit/receive antennas Moving speed 3 km/hr, 30 kmfhr, 120 km/hr Data modulation QPSK and 16QAM Channel coding Turbo code with softdecision deciding Coding rate M and 1/3 Equalizer LMMSE Group feedback One feedback per 1, 12 and 25 subuarriors Feedback error None (Assumed ideal) Feedback delay 2 and 6 TTIs -11- Channel Estimation Ideal channel estimation . Table 1. [00561 Sinulati on Results and Discussions [0057] Figure I shows the performance of MIMO preceding for a TU6 channel model and vehicle speed at 8kSm/hr. The performance of MIMO preceding with group feedback of different group sizes is compared. No group feedback is feedback per subcarrier which requires the highest feedback overhead. Group feedback uses one feedback for every L subcarriers, About 0,3 dB degradation is observed for group feedback using one feedback per 12 subcarriers with respect to the performance of no group feedback, i.e., L=1 About 0.8 dB degradation in performance is observed for group feedback using one feedback per 25 subcarriers with respect to no group feedback. [0058] In addition the performance of MIMO preceding with and without quantization is compared in Figure 1. With differential feedback that uses 2 bits per feedback group, about 0.3 dB degradation results from quantization for all group feedback sizes, L=1, 12 and 25 subcarriers is observed. The feedback was updated every TTI and was reset every 10 TTIs. [0059 Figure 2 shows the performance of MIMO precoding using group feedback and codebook quantization for an SCVIE-O channel and vehicle speed at 3 km/hr. About 0.1 dB degradation is observed for group feedback using one fbedback per 12 subcarriers with respect to the performance ofno group feedback, i.e., L=1. About 0.2 dB degradation is observed for group feedback using one feedback per 25 subcarriers with respect to no group feedback. In addition about 0.3 dB degradation due to quantization that uses 2 bits per feedback group is observed. [0060] Figure 3 shows the performance comparison for MIMO preceding using differential and non-differential feedback. The performance of combined differential and non-differential feedback that uses nxed 2 bits/Sbits scheme is compared against non-differential feedback using 3 bits. Combined differential and non-differential feedback uses 2-bit quantization with 3-bit quantization at each resetting period, -12- 100611 It is observed that the performance of differential feedback using fewer bits (2 bits) with proper resetting interval for differential processing is similar to the performance of non-differential feedback using fil feedback and more bits (3 bits), The combined differential and non-differential feedback can reduce the feedback overhead by as much as 33% as compared to feedback overhead of non-differential feedback, depending on the iteration interval and reset period. About 0.3-0.4 dB degradation in performance for preceding using quantization with respect to ideal precoding/TxBF with no quantization. [0062] Figure 4 shows the performance of MXMO precoding using differential feedback with resetting. It is shown that the performance of differential feedback every TTI with proper resetting may improve the performance by 2 dB. This is because the preceding error due to qnantiz9ation may accumiate or propagate for differential feedback. The resetting process corrects the error, thus improving the performance. [0063] The performance of differential feedback with different resetting intervals of N=10, 20, 30 and 50 TTIs are compared. Performance degradation is negligible; about 0.1 dB degradation in performance is observed with the longest resetting interval of 50 TTIs. Note that this does not account for the effects of possible feedback bit errors; however, we believe that such errors will be rare because of error protection. [00641 Figure 5 shows the performance of MffMO precoding using differential feedback with feedback delay for an SCME-C channel and vehicle speed 8 km/h. The combined performance degradation for 2-bit quantization and feedback delay is about 0.3 dB for feedback delay of 2 TTIs and about 0.4 dB for feedback delay of 6 TTIs with respect to no quantization and no feedback delay. [00651 Figure 6 shows the performance of MIMO preceding using differential feedback with feedback delay for an SOME-C channel and vehicle speed 120 km(h. It is shown that about 0.6 dB degradation results from 2 TTI feedback delay and about 15 dB degradation results from 6 TTI feedback delay with respect to the performance of no feedback delay. When compared to the performance of ideal precoding with no quantization and no feedback, the -13performance of differentialfeedback has about 1.7 dB and 2.7 dB degradation for combined quantization and feedback delay of 2 TTs and 6 TTIs respectively. [00661 Figure 7 shows the performance of MIMO preceding using non-. differential feedback for an SCME-C channel and 120 km/h. It is shown that the performance degrades about 0.6 dB for 2 TTI feedback delay and about 2 dB for 6 TTI feedback delay as compared to the performance of no feedback delay. When compared with the performance of ideal preceding with no quantization and no feedback, the performance of diierential feedback has about 0.7 dB and 2.2 dB degradation for combined quantization and feedback delay of 2 TTIs and 6 TTIs correspondingly. A shorter feedback delay is obviously preferable for such high speed channels to reduce the performance loss due to speed. [0067] MIMO precoding using differently feedback, non-differential, and group feedback can be applied to uplink or downt-tk MIMO for SC-FDMA or OFDMA air interfaces. The following shows the differential feedback work for uplink MIMO with a SC-FDMA air interface. 100682 These techniques may be extended to any number of antennas greater than one. 100691 Architecture [0070] Figure 8A is a block diagram of a transmitter 800 for a DCW configuration of uplink MIMO using precoding with dual transmit chains in accordance with the present invention. In the case of an SOW, the coded data is split into parallel streams, each with a different modulation The transmitter 800 may be an eNodeB or a base station, (i.e., the eNodeB in LTE terminology). [00711 Referring to Figure 8A, the transmitter 800 includes a demultiplexer 810, a plurality of channel encoders 8151-8I5, a plurality of rate matching units 820-820,, a plurality of frequency interleavers 8 2 5 r 8 2 5 a, a plurality of constellation mapping umits 88-830S, a plurality of fast Fourier transform (FFT) units 8351-835n, a precoder 840, a subcarer mapping unit 845, a plurality of multiplexers 8501-850n, a plurality of inverse FFT (IFFT) units 855r,-855o, a plurality of cyclic prefix (CP) insertion units 8601-860t, a plurality of antennas 865z-865n and a preceding matrix generator 876. It should be noted -14that the configuration of the transmittr 800 is provided as an example, not as a limitation, and the processing may be performed by more or less components and the order of processing may be switched. [0072] Transmit data 805 is first demultiplexed into a plurality of data streams 8121-812a by the demultiplexer 810. Adaptive modulation and coding (AMC) may be used for each of the data streams 8121-812., Bits on each of the data streams 8 1 2 1- 8 1 2 n are then encoded by each of the channel encoders 8151 815, to generate encoded bits 8181-818n, which axe then punctured for rate matching by each of the rate matching units 820a-820n. Alternatively, multiple input data streams may be encoded and punctured by the channel encoders and rate matching units, rather than parsing one transmit data into multiple data streams. [0078] The encoded data after rate matching 8221-822, is preferably interleaved by the inteieavers 825i-825n.. The data bits after interleaving 828 828n are then mapped to symbols- 8321-832k by the constellation mapping units 830-830 in accordance with a selected modulation scheme. The modulation scheme may be binary phase shift keying (BPSIQ, quadrature phase shift keying (QPSK), SPSK, 16 quadrature amplitude modulation (QAM), 64 QAM, or similar modulation schemes, Symbols 3 2 1- 8 3 2 . on each data stream are processed by the FFT units 885i-8852, which outputs frequency domain data 8381-838. [0074] The precoding matrix generator 875 uses non-differential or differential feedback bits, (or feedback channel metrics), to generate a set of precoding weights 880 (i.e., a precoding matrix), which are fed to the precoder 840 for performing preceding on the frequency domain data streams 8 8 8

L-

8 38 . [0075] Figuxes SB and SC show details of the precoding matrix generator 875 of the transmitter 800 of Figure 8A. [0076] If the feedback bits 870 include non-difbrential feedback bits 870', the precoding matrix generator 875 may be confgured as the preceding generator 875' shown in Figure 8B. The precoding matrix generator 875' includes a feedback bits to full preceding matrix mapping unit 890 that translates the non -15differential feedback bits 870' into a fil pre coding matrix 880' (J) using a non differential codebook 888. f0077] If the feedback bits 870 include differential feedback bits 870", the precoding matrix generator 875 may be configured as the precoding matrix generator 875" shown in Figure 80. The preceding matrix generator 875" includes a feedback bits to delta preceding matrix mapping unit 894 that translates the differential feedback bits 870 into a delta precoding matrix 896 (A') using a differential codebook 892. The delta preceding matrix 896 is represented by AO and Aib. The precoding matrix generator 875" further includes a full preceding matrix generation and update unit 898 that translates the delta precoding matrix 896 to a full precoding matrix 880" (J), which is represented by 9 and j. 100781 Referring back to Figure BA, the precoder 840 applies the weights to each stream of frequency domain data 888-8388 similar to spatial spreading or beamforming, and outputs precoded data streams 8 4 2r 8 4 2., The subcarrier mapping unit 845 maps the preceded data streams 842a-842, to the subcarriers that are assigned for the user. The subcarrier mapping may be either distributed subcarrier napping or localized subcarrier mapping. [0079] The subcarrier mapped data 84218j- 4 2 n is multiplexed with pilots 849 by the multiplexers 8501-850", the outputs 852i-85 2 of 'which are then processed by the IFFT units 855-855.. The IFFT units 855,-855&, output time domain data 858-858.. A CP is added to each time domain data stream 8581 858n by the OP insertion units 8601-80Sk The time domain data with CP 862t 862. is then transmitted via the antennas 866-865.. [00801 Figure 9A is a block diagram of a receiver 900 that receives and processes signals transmitted by the transmitter 800 of Figure SA in accordance with the present invention. A single decoder may be used in the SOW case. The receiver 900 may be a WTRU. [00811 The precoder matrix codeword index is assumed to be fed back from the base station, (i.e., the eNodeB in LTE terminology), to the WTRU. -16- [00823 The receiver 900 includes a plurality of antennas 905-90O6, a plurality of CP removal units .91O-910, a plurality of FFT units 915i-915m, a channel estimator 920, a subcarrier demappiug unit 925, a MIMO decoder 930, a plurality of IVFT units 93 5 1- 9 8 5 n, a plurality of data demodulators 9401-940., a plurality of deinterleavers 9456-945n, a plurality offorward. error correction (FEC) units 9501-95On, a spatial deparser 955 and a feedback generator 960. The MIMO decoder 930 may be a ininimum mean square error (MMSE) decoder, an MMSE-successive interference cancellation (SIC) decoder, a maximum likelyhood (ML) decoder, or a decoder using any other advanced techniques for.MIMO. [0083] Still referring to Figure 9A, the CP removal units 9 10--9Y10, remove a CP from each of the data streams 9081-908n received by the antennas 9051 905.. After CP removal, the processed data streams 9121-912. output by the CP removal units 9 1 0 1n910. are converted to frequency domain data 9181-918n by the FFT units 9151-915.. The channel estimator 920 generates a channel estimate 922 from the frequency domain data 918i-918% using conventional methods. The channel estimation is performed on a per subcarrier basis. The snbearrer demapping unit 925 performs the opposite operaUion which is performed at the transmitter 800 of Figure 8A. The subcarrier demapped data 9281-9281 is then processed by the MIMO decoder 930. [0084] After MIMO decoding, the decoded data 932,-932. is processed by the IFFT units 9 351- 9 85n for conversion to tine domain data 938j-988. The time domain data 9381-938, is processed by the data demodulators 9 4 0 i- 9 4 0n to .generate bit streams 942i- 9 4 2 ,. The bit streams 9 4 2 -9 4 2 . are processed by the deinterleavers 944-9 4 5 n, which perform the opposite operation of the interleavers 825-825n of the transmitter 800 of Figure 8A. Each of the deinterleaved bit streams 9481-948. is then processed by each of FEC units 9501 950.. The data bit streams 9521-952, output by the FEC units 9 50i 9 56 are merged by the spatial de-parser 955 to recover data 962. The feedback generator generates non-differential or differential feedback bits, which are fed back to the precoding matrix generator 875 of the transmitter 800. -17- [0085) Figures 9B and 90 show details of the feedbackgenerator 960 of the receiver 900 of Figure 9A. [0086] If the feedback bits 870 include non-differential feedback bits 870', the feedback generator 960 may be conilgured as the feedback generator 960' shown in Figure 9B. The feedback generator 960' includes a preceding matrix generator 1005', which outputs a fall preceding matrix 1010 (J) in the formof its parameters 0 and . The full preceding matrix 1010 is fed to a feedback bit generator 1020', which uses a non-differential codebook 1015 to generate non differential feedback bits 870'. [0087] If the feedback bits 870 include differential feedback bits 870", the feedback generator 960 may be configured as the feedback generator 960" shown in Figure 90. The feedback generator 960" includes a precoding matrix generator 1005", which outputs a delta preceding matrix 1012 (Al) in the form of its parameters AO and Ab. The delta preceding matrix 1012 is fed to a feedback bit generator 1020", whichuses a differential codebook 1018 to generate differential feedback bits 870". [0088} Figures IA and 1OB show different embodiments of the preceding matrix generator 1005'ned in the feedback generator 960' of Figure 9B. In one embodiment, the preceding matrix generator 1005' generates a full preceding matrix 1010'used to generate non-differential feedback bits based on Equations (1) and (6b). In another embodiment, the precoding matrix generator 1005' generates a full preceding matrix 1010" used to generate non-differential feedback bits based on Equations (2), (9) and (10). 10089] Figures 10C and 10D show different embodiments of the preceding matrix generator 1005" used in the feedback generator 960" of Figure 9C. In one embodiment, the preceding matrix generator 1005" generates a delta preceding matrix 1012' used to generate differential feedback bits based on nations (2), (12), (15) and (16). In another embodiment, the preceding matrix generator 1005" generates a delta precoding matrix 1012" used to generate differential feedback bits based on Equation (17). -18- [00901 Precoding [00911 The precoding is based on transmit beamforming (TxBF) using, for example, eigen-beamforming based on SVD. While SVD is optimal, other algorithms may be used by the Node B. [0092) As previously shown by Equation (1), the channel matrix is decomposed using an SVD or equivalent operation as H = UDV", where H is the channel matrix. The preceding for spatial multiplexing, beamforming, and the like, can be expressed as x t Ts, Equation (20) where s is the data vector and T is a generalied precoding matrix or transform matrix. In the case when transmit eigen-beamforming is used, the preceding or transform matrix T is chosen to be a beamforming matrix V, which is obtained from the SVD operation above, i.e., T = V. Alternatively, the precoding or transform matrix T is chosen from a codebook or quantization. The selection of the codeword among codebook or quantization for precoding matrix T is based on some predetermined criterion, such as SINR, mean square error (MSE), channel capacity, and the like. Based on estimated channel matrix H, the precoding matrix among all candidate precoding matrices which has highest metrics, such as highest SNIR, largest channel capacity or smallest MSE is selected. Alternatively, based on SVD operation, the codeword or piecoding matrix among all candidate preceding matrices in codebook that is the best quantization of the matrix V is selected. This is similar to eigen-beamnforning for OFDMA, modified to apply to SQ-FDMA. [00931 Because the SVD operation results in orthogonal streams, the eNodeB can use a simple linear MMSE (LMMSE) receiver. It can be expressed as R = R.

5 5 " (fR1 " +R%)-', Equation (21) where R is a receive processing matrix, R,, and -R, are correlation matrices and fis an effective channel matrix which includes the effect of the V matrix on the -19estimated channel response. In Figure SA, the precoder 840 in the eNodeB (i.e., transmitter 800) produces the effective channel matrix at the WTRTJ using the last quantized precoder matrix sent from the eNodeB to the WTRU. [0094 Feedback {0095] An approach to feeding back the preceding matrix employs a codebook-based MIMO preceding scheme using combined differential and non differential feedback as described in the early section. [0096) This section presents selected simulation results for SU-MIMO. A comparison between SU-MIMO and SIMO is discussed first, followed by a comparison of the performance for single and double codeword SU-MIMO. [0097] Simulation parameters [098] The simulation paarneters assumed axe provided in Table 1. The achievable throughputs for various selections of the MOS for each spatial stream are provided in Table 2 below. Azbievable Spectral MCS Data Rate Efficiency (Mps) (bps/H.) 16QAM r7/8- 16QAM r3/4 19.9680 3.99 16QAM r7/8- 16QAM r1/2 16.8960 3.38 16QAM r7/8- 16QAM r1/3 14.8480 2.97 16QAM r/6 - QPSK ri/S 1.1.08 2.22 16QAM r5/6 - QPSK r1/2 10.752 2.15 16QAM r3/4 - QPSK r/6. 10.24 2.05 16QAM r3/ 2 - QPSKrt/3 8.192 1.64 16QAM r1/2 - QP$K r1/6 7.168 1.43 16QAM r1/3 - QPSK r18 4,864 0.97 16QAM r1/4 - QPSK r1/8 3.840 07 Table 2. [099] It is worth noting that the maximum achievable throughput using a double codeword and practical code rates in 5 MHz is 19.968 Mbps, which scales to 79.87 Mbpn in a 20 VUIs bandwidth, and has a spectral efficiency of 4 bps/Hz. SIMO, on the otherhand, is limited to 10.76 Mbps in 6 MHz, a spectral efficiency of 2.15. Therefore, SU-MIMO can almost double the uplink data rate compared with SIMO. -20- 101001 Comparisn of SU-MIMO to SIMO [01011 Figure 11 shows a comparison of double codeword performance for.SUY MIMO to SIMO for the high data throughput SNR regions. When the SNR is 24 dB the maximum achievable throughput is approximately 19 Mbps, and when the SNR is greater than 26 dB the achievable throughput is approdmately 19.97 Mbps. From this comparison it is worth noting that using SIMO the maximum achievable throughput is 10.5 Mbps at an SNR of 20 dB. [0102] Comparison of SU-MIMQ with sinle and double codewords (0103) This section presents a comparison of the performance for single and double codewords using uplink precoding MIMO for two antennas at the WTRU and eNodeB with the SOME-C channel. Because HARQ was not simulated, the same code rate was used for both SCW and DOW in order to compare them fairly. Also, it is impractical to use the same modulation for SCW for both streams when using preceding, so only combinations .of QPSK and 16QAM are shown. Therefore, the higher throughput achievable with DOW is not shown. [0104] Figure 12 shows a comparison ofthe performance for single and double codewords using uplink preceding MIMO for two antennas at the WTRU and eNodeB with an SOME-C channeL (0105] The DCW achieves a higher throughput at lower SNRs, while the opposite is true at higher SNRs. The SCW performs better than DOW. The difference is more pronounced at the highest data rates where a S3dB difference can be seen. Eventually, since equal modulation and coding was used, both schemes reach the same maximum throughput, almost 14 Mbps in 5 MHz for the highest MCS simulated. [0106 The reason that DOW performs beer at lower SNR is because the upper eigen-mode has higher SNR than the total system SNR. Therefore at low SNR that stream contributes some successful transmissions while the lower stream generally does not. However, at higher SNR the lower stream still has relatively high BLER which tends to reduce the total throughput for DOW. But, in the case of SOW, the upper stream protects the lower stream because the -21coding covers both streams. This results in anu overall lower BL'R) for SCW at higher SNTs.

[0107] From these results it may be concluded that very high uplink spoetral efficiency, about 2.8 bpa/li can be achieved using either method. However, DOW can achieve a higher spectral efficiency, about 4 bps/Hz, because it can use 16QAM with different code rates on each streak, whereas SOW must use a single code rate and different modulations. [0108) In summary, uplink SU-MIMO for SO-FDMA according to the preferred embodiments achieve the following: 1) Precoding at the UE can be based on SVD or a comparable algorithm performed at the eNadeB For an SCUM-N channel the codebook can be based on channel averages taken over several, e.g. six adjacent RBs. 2) Feedback of the precoding matrix index can be performed efficiently using combined differential and non-differential feedback. Representative feedback parameters are 2 bits every 6 RBs sent every 6 TTIs, or a maxi-mum of 1338 bps for 24 RBs in 5 MHz. Since the equivalent maximum data rate is 19,968 Mbps, the feedback efficiency is very high. 3) Simulations showed that SU-MIMO can almost double (186 %) the uplink data rate compared with SIMO. [0109] Embodiments 1. A method of providing precoding feedback in a multiple input mualtiple output (MIMO) wireless communication systemincluding a receiver and a transmitter, the method comprising the receiver transmitting either non-differential feedback bits or differential feedback bits; and the transmitter updating a first preceding matrix based on the feedback bits and preceding a plurality of frequency domain data streams using the first preceding matrix. 2, The method of embodiment 1 further comprising: the transmitter transmitting a plurality of time domain data streams, each time domain data stream including a cyclic prefix (CP); ' -22the receiver receiving the time domain data streams; the receiver removing the CPs from the time domain data streams to generate a plurality of processed data streams; the receiver conrverting the processed data streams to freqaeay domain data; the receiver perfonning channel estimation on the frequency domain data to generate a channel estimate; the receiver generating a second preceding matrix based on the channel estimate; and the receiver generating and transmitting feedback bits based on the second preceding matrix. 3. The method of embodiment 2 wherein the second preceding matrix is a delta preceding matrix and the feedback bits are differential feedback bits. 4. The method of embodiment 2 wherein the second preceding matrix is a full preceding matrix and the feedback bits are non-ditferential feedback bits. 5. The method of embodiment 4 wherrin non-differential feedback bits are generated by using a Jacobi rotation to perform matrix diagonalization On at least one of a clinnel response matrix and a channel correlation matrix associated with the channel estimate. 6. The method as in any one of embodiments 1-5 wherein the feedback bits are non-differential feedback bits, the method further comprising: the transmitter mapping the non-differential feedback bits to a fall preceding matrix by using a non-differential codebook. 7. The method as in any one of embodiments 1-5 wherein the feedback bits are differential feedback bits, the method further comprising; the transmitter mapping the non-differential feedback bits to a delta preceding matrx by using a differential codebook; and the tranntitter generating a full preceding matrix based on the delta preceding matrix. 8. The method as in any one of embodiments 1-7 wherein the receiver is a wireless transmit/receive unit (WTRUL -22-.

Claims (10)

1. A method of providing precoding feedback including: receiving a plurality of feedback bits; updating a first precoding matrix based on the feedback bits, wherein the feedback bits are 5 differential feedback bits; precoding a plurality of frequency domain data streams using the first precoding matrix; mapping the differential feedback bits to a delta precoding matrix by using a differential codebook; and generating a full precoding matrix based on the delta precoding matrix. 10

2. A receiver for providing feedback to a transmitter for updating a first precoding matrix used by the transmitter to precode a plurality of frequency domain data streams, the receiver including: a channel estimator configured to generate a channel estimate by performing a channel estimation on frequency domain data associated with a plurality of time domain data streams transmitted by the transmitter; and 15 a feedback generator electrically coupled to the channel estimator, the feedback generator configured to generate feedback bits for transmission to the transmitter based on the channel estimate, wherein the feedback bits are differential feedback bits, the feedback generator including: a precoding matrix generator configured to generate a delta precoding matrix based on the channel estimate; and 20 a feedback bit generator electrically coupled to the delta matrix generator, the feedback bit generator being configured to generate and transmit feedback bits based on the delta precoding matrix.

3. A transmitter that performs precoding based on feedback provided by a receiver, the feedback being generated based on a plurality of time domain data streams that the receiver receives 25 from the transmitter, the transmitter including: a precoding matrix generator configured to receive feedback bits from the receiver and update a precoding matrix based on the feedback bits, wherein the feedback bits are either non differential feedback bits or differential feedback bits; and 25 a precoder electrically coupled to the precoding matrix generator, the precoder being configured to precode a plurality of frequency domain data streams using the precoding matrix, the precoder including: a feedback bits to delta precoding mapping unit for mapping differential feedback 5 bits to a delta precoding matrix; and a full precoding matrix generation and update unit for generating and updating a full precoding matrix based on the delta precoding matrix, wherein the precoder uses the full precoding matrix to precode the frequency domain data streams.

4. The transmitter of claim 3 wherein the precoder includes: 10 a feedback bits to full precoding mapping unit for mapping non-differential feedback bits to a full precoding matrix, wherein the precoder uses the full precoding matrix to precode the frequency domain data streams.

5. The transmitter of claim 3 wherein the receiver is a wireless transmit/receive unit (WTRU).

6. The transmitter of claim 3 wherein the transmitter is an evolved Node-B (eNodeB). 15

7. The transmitter of claim 3 wherein the transmitter is a base station.

8. The method of claim 1 and substantially as hereinbefore described with reference to the accompanying figures.

9. The receiver of claim 2 and substantially as hereinbefore described with reference to the accompanying figures. 20

10. The transmitter of claim 3 and substantially as hereinbefore described with reference to the accompanying figures.