Title of Invention  INTERFERENCE SUPPRESSION WITH VIRTUAL ANTENNAS 

Abstract  A receiver suppresses cochannel interference (CCI) from other transmitters and intersymbol interference (ISI) due to channel distortion using 'virtual' antennas. The virtual antennas may be formed by (1) oversampling a received signal for each actual antenna at the receiver and/or (1) decomposing a sequence of complexvalued samples into a sequence of inphase samples and a sequence of quadrature samples. In one design, the receiver includes a preprocessor, an interference suppressor, and an equalizer. The preprocessor processes received samples for at least one actual antenna and generates at least two sequences of input samples for each actual antenna. The interference suppressor suppresses cochannel interference in the input sample sequences and provides at least one sequence of CCIsuppressed samples. The equalizer performs detection on the CCIsuppressed sample sequence(s) and provides detected bits. The interference suppressor and equalizer may be operated for one or multiple iterations. 
Full Text  FORM 2 THE PATENTS ACT, 1970 (39 of 1970) & The Patents Rules, 2003 COMPLETE SPECIFICATION (See section 10, rule 13) "INTERFRENCE SUPPRESSION WITH VIRTUAL ANTENNAS" QUALCOMM Incorporated of 5775 Morehouse drive Sandiego, California 921211714, United States of America, State of Incorporation: Delaware. The following specification particularly describes the invention and the manner in which it is to be performed. WO 2006/055797 PCT/US2005/041856 2 INTERFERENCE SUPPRESSION WITH VIRTUAL ANTENNAS [0001] This application claims the benefit of provisional U.S. Application Serial No. 60/629,656, entitled "MIMO based SAIC Algorithms for GSM/GPRS," filed November 19, 2004, assigned to the assignee of the present application, and incorporated herein by reference in its entirety for all purposes. BACKGROUND I. Field [0002] The present invention relates generally : to communication and more specifically to a receiver in a communication system. II. Background [0003] In a communication system, a transmitter processes data to generate a modulated signal and transmits the modulated signal on a frequency band/channel and via a communication channel to a receiver. The transmitted signal is distorted by the communication channel, corrupted by noise, and further degraded by cochannel interference, which is interference from other transmitters transmitting on the same frequency band/channel. The receiver receives the transmitted signal, processes the received signal, and attempts to recover the data sent by the transmitter. The distortion due to the communication channel, the noise, and the cochannel interference all hinder the receiver's ability to recover the transmitted data. [0004] There is therefore a need in the art for a receiver that can effectively deal with cochannel interference and channel distortion. SUMMARY [0005] A receiver capable of suppressing cochannel interference (CCI) from other transmitters and intersymbol interference (ISI) due to channel distortion using "virtual" antennas is described herein. The virtual antennas may be formed by (1) oversampling a received signal for each actual antenna at the receiver and/or (1) decomposing a WO 2006/055797 PCT/US2005/041856 3 sequence of complexvalued samples for each actual antenna into a sequence of inphase samples and a sequence of quadrature samples, where the inphase and quadrature samples are for the real and imaginary parts, respectively, of the complexvalued samples. If the receiver is equipped with Nant actual antennas, where Nant > 1, then 2Nant virtual antennas may be obtained via real/imaginary decomposition, Nant Nos virtual antennas may be obtained via Nos times oversampling, and 2Nam Nos virtual antennas may be obtained via real/imaginary decomposition and Nos times oversampling. [0006] In an embodiment, the receiver includes a preprocessor, an interference suppressor, and an equalizer. The preprocessor processes the received samples for at least one actual antenna and generates at least two sequences of input samples for each actual antenna. Each input sample sequence corresponds to one virtual antenna. The preprocessor performs processing pertinent for the modulation scheme used for transmission, e.g., phase rotation for Gaussian minimum shift keying (GMSK) used in a Global System for Mobile Communications (GSM) system. The interference suppressor suppresses cochannel interference in the input sample sequences and provides at least one sequence of CCIsuppressed samples. The equalizer performs detection on the CCIsuppressed sample sequence(s) and provides detected bits. [0007] In an embodiment, the interference suppressor includes a channel estimator, a signal estimator, a computation unit, and a multipleinput multipleoutput (MTMO) filter. The channel estimator derives at least one channel estimate based on the input sample sequences. The signal estimator derives ai. least one desired signal estimate based on the at least one channel estimate. The computation unit computes weights used for cochannel interference suppression. The MTMO filter filters the input sample sequences with the weights and provides the CCIsuppressed sample sequence(s). [0008] In an embodiment, the equalizer includes a channel estimator and a detector. The channel estimator derives at least one improved channel estimate based on the at least one CCIsuppressed sample sequence from the interference suppressor. The detector performs detection on the CCIsuppressed sample sequence(s) with the improved channel estimate(s) and provides the detected bits. [0009] Other embodiments of the interference suppressor and equalizer are described below. Various other aspects and embodiments of the invention are also described in further detail below. WO 2006/055797 4 PCT/US2005/041856 BRIEF DESCRIPTION OF THE DRAWINGS [0010] The features and nature of the present invention will become more apparent from the detailed description set forth below when taken in conjunction with the drawings in which like reference characters identify correspondingly throughout. [0011] FIG. 1 shows a transmitter and a receiver in a wireless communication system. [0012] FIG. 2 shows exemplary frame and burst formats in GSM. [0013] FIG. 3 shows a conventional demodulator and receive (RX) data processor for GSM. [0014] FIG. 4 shows a demodulator capable of performing cochannel interference suppression using virtual antennas. [0015] FIG. 5 shows two sample sequences obtained via 2x oversampling. [0016] FIG. 6A shows a model for two transmitters with binary phase shift keying (BPSK). [0017] FIG. 6B shows MIMO models for two transmitters with BPSK. [0018] FIG. 7A shows spacetime processing for interference suppression with virtual antennas. [0019] FIG. 7B shows a MIMO filter that performs spacetime processing on two complexvalued input sample sequences for cochannel interference suppression. [0020] FIG. 7C shows a finite impulse response (FIR) filter within the MIMO filter. [0021] FIG. 8 shows a demodulator that suppresses cochannel interference using virtual antennas. [0022] FIG. 9 shows a demodulator that suppresses cochannel interference using virtual antennas and performs detection with noise decorrelation. [0023] FIG. 10 shows a demodulator that suppresses interference using bit pruning. [0024] FIG. 11 shows a demodulator that suppresses interference using reencoded bits [0025] FIG. 12 shows a demodulator and an RX data processor that perform iterative interference suppression and decoding. DETAILED DESCRIPTION [0026] The word "exemplary" is used herein to mean "serving as an example, instance, or illustration." Any embodiment or design described herein as "exemplary" WO 2006/055797 5 PCT/US2005/041856 is not necessarily to be construed as preferred or advantageous over other embodiments or designs. [0027] The receiver with virtual antennas may be used in various communication systems. For clarity, the receiver is specifically described below for GSM. [0028] FIG. 1 shows a block diagram of a transmitter 110 and a receiver 150 in a wireless communication system. Transmitter 110 may be a base station or a wireless device, and receiver 150 may also be a wireless device or a base station. At transmitter 110, a transmit (TX) data processor 120 receives, formats, encodes, and interleaves data based on a coding and interleaving scheme and provides a stream of input bits to a modulator 130. For GSM, modulator 130 performs GMSK modulation on the input bits and provides a GMSK modulated signal (or simply, a GMSK signal). GMSK is a continuous phase modulation scheme used in GSM and is described below. A transmitter unit (TMTR) 132 conditions (e.g., filters and amplifies) the GMSK signal and generates a radio frequency (RF) modulated signal, which is transmitted via an antenna 134 to receiver 150. [0029] At receiver 150, an antenna 152 receives the RF modulated signal from transmitter 110 and RF modulated signals from other transmitters in the GSM system. Antenna 152 provides a received GMSK signal to a receiver unit (RCVR) 154. Receiver unit 154 conditions and digitizes the received GMSK signal and provides received samples. A demodulator 160 processes the received samples and provides detected bits, as described below. An RX data processor 170 processes (e.g., deinterleaves and decodes) the detected bits and recovers the data sent by transmitter 110. The processing by demodulator 160 and RX data processor 170 is complementary to the processing by modulator 130 and TX data processor 120, respectively, at transmitter 110. [0030] Controllers 140 and 180 direct operation at transmitter 110 and receiver 150, respectively. Memory units 142 and 182 provide storage for program codes and data y used by controllers 140 and 180, respectively. [0031] FIG. 2 shows exemplary frame and burst formats in GSM. The timeline for downlink transmission is divided into multiframes. For traffic channels used to send userspecific data, each multiframe includes 26 TDMA frames, which are labeled as TDMA frames 0 through 25. The traffic channels are sent in TDMA frames 0 through 11 and TDMA frames 13 through 24 of each multiframe. A control channel is sent in WO 2006/055797 PCT/US2005/041856 6 TDMA frame 12. No data is sent in the idle TDMA frame 25, which is used by the wireless devices to make measurements for neighbor base stations. [0032] Each TDMA frame is further partitioned into eight time slots, which are labeled as time slots 0 through 7. Each active wireless device/user is assigned one time slot index for the duration of a call. Userspecific data for each wireless device is sent in the time slot assigned to that wireless device and in TDMA frames used for the traffic channels. [0033] The transmission in each time slot is called a "burst" in GSM. Each burst includes two tail fields, two data fields, a training sequence (or midamble) field, and a guard period (GP). The number of bits in each field is shown inside the parentheses. GSM defines eight different training sequences that may be sent in the training sequence field. Each training sequence contains 26 bits and is defined such that the first five bits (labeled as 'A') are repeated and the second five bits (labeled as 'B') are also repeated, as shown in FIG. 2. Each training sequence is also defined such that the correlation of that sequence with a 16bit truncated version of that sequence (with parts 'B\ 'C\ and 'A') is equal to (a) sixteen for a time shift of zero, (b) zero for time shifts of ±1, ±2, ±3, ±4, and ±5, and (3) a zero or nonzero value for all other time shifts. [0034] To generate a GMSK signal, modulator 130 receives input bits a from TX data processor 120 and performs differential encoding on the input bits to generate code symbols c. One new code symbol is generated for each new input bit. Each input bit and each code symbol have a real value of either +1 or —1. Modulator 130 further filters each code symbol with a Gaussian lowpass filter to generate a frequency pulse having a duration of approximately four sample periods (4T). Modulator 130 integrates the frequency pulses for the code symbols to generate a modulating signal and further modulates a carrier signal v/ith the modulating signal to generate the GMSK signal. [0035] The GMSK signal has a complex representation but may be approximated as follows: where (X) denotes a convolution operation; Pi denotes the fth pulse shaping function; and di denotes the input symbols for pulse shaping function/?/. WO 2006/055797 7 PCT/US2005/041856 Equation (1) indicates that the complex GMSK signal may be expressed as a sum of amplitudemodulated signals. Each amplitudemodulated signal is generated by convolving a pulse shaping function Pi with its input symbols di. For GMSK, there are eight pulse shaping functions Pi through pi, with Pi being the dominant pulse shaping function that is much larger than the other seven pulse shaping functions. The input, symbols dt for each pulse shaping function p, are derived from the input bits a based on a specific transformation associated with function pi,,. For example, the input symbols do for the dominant pulse shaping function Pi may be expressed as: where a(t) is the input bit for sample period t, j = Vl, and d0(t) is the input symbol for the dominant pulse shaping function for sample period t. Equation (2) indicates that the input symbols do for the dominant pulse shaping function are generated by rotating the input bits a by successively larger phases, or 0° for a(t), then 90° for a(t + l), then 180° for a(t + 2), then 270° for a{t + 3), then 0° for a{t + 4), and so on. [0036] FIG. 3 shows a demodulator 160a and an RX data processor 170a, which are conventional designs for demodulator 160 and RX data processor 170, respectively, at receiver 150 in FIG. 1. Within demodulator 160a, an RX filter 312 filters the received samples rn from receiver unit 154 and provides intermediate samples r. The intermediate samples may be expressed as: where hc is the impulse response of the wireless channel from transmitter 110 to receiver 150; Pi ®hc is the z'th effective pulse shaping function; vr is the cochannel interference from other transmitters; and nf is the noise at the receiver. [0037] A GMSKtoBPSK rotator 314 performs phase rotation on the intermediate samples r and provides input samples z. The phase rotation may be expressed as: XO 2006/055797 8 PCT/US2005/041856 z(t) = j'r(t) , Eq(4) where r(t) is the intermediate sample for sample period t; and z{t) is the input sample for sample period t. Rotator 314 rotates the intermediate samples by successively more negative phases, or 0° for r(t), then 90° for r(t +1), then 180° for r(t + 2), then 270° for r(t + 3), then 0° for r(r + 4), and so on. The phase rotation results in the input symbols d0{t) for the effective dominant pulse shaping function p0 hc being equal to the input bits provided to modulator 130, or d0(t)  f • d0(t) = a(t). [0038] To simplify the receiver design, the GMSK signal may be approximated as a BPSK modulated signal that is generated with just the dominant pulse shaping function. The input samples may then be expressed as: Eq (5) where h = p0®hc is the effective channel impulse response; v is a rotated version of the cochannel interference vr; and n is the total noise, which includes a rotated version of the noise nr and components of the other pulse shaping functions. The GMSKtoBPSK approximation in equation (5) is a reasonably good approximation since the dominant pulse shaping functionpo is much larger than the other pulse shaping functions. [0039] An equalizer 350 performs equalization on the input samples z to remove intersymbol interference caused by multipath in the wireless channel. For the design shown in FIG. 3, equalizer 350 includes a channel estimator 360 and a detector 370. Channel estimator 360 receives the input samples z and the training sequence xls and derives an estimate of the effective channel impulse response h, The effective channel impulse response estimate h is approximately equal to the dominant pulse shaping function convolved with the actual channel impulse response, or [0040] Detector 370 receives the input samples z and the channel estimate h and performs detection to recover the input bits a. Detector 370 may implement a WO 2006/055797 PCT/US2005/041856 9 maximum likelihood sequence estimator (MLSE) that determines a sequence of bits that is most likely to have been transmitted given the input sample sequence z and the channel estimate h. The MSLE uses a Viterbi algorithm with 2L_1 states, where L is the length of the channel estimate it. Detection with MLSE for GSM is well known in the art and not described herein. Detector 370 provides detected bits x,ie,, which are hard decision estimates of the input bits a sent by transmitter 110. [0041] Within RX data processor 170a, a soft output generator 380 receives the detected bits Xdet and the input samples z and generates soft decisions that indicate the confidence in the detected bits. Soft output generator 380 may implement an Ono algorithm that is well known in the art. A deinterleaver 382 deinterleaves the soft decisions in a manner complementary to the interleaving performed by transmitter 110. A Viterbi decoder 384 decodes the deinterleaved soft decisions and provides decoded data c, which is an estimate of the traffic data provided to TX data processor 120 at transmitter 110. [0042] FIG. 4 shows a demodulator 160b capable of performing cochannel interference suppression using virtual antennas. Receiver unit 154 may digitize the received GMSK signal at twice the sample rate and provide 2x oversampled received samples rrx. Within a preprocessor 410, an RX filter 412 filters the received samples and provides a sequence of "early" samples r\ and a sequence of "late" samples r%. RX filter 412 may be a polyphase filter or some other type of filter. A GMSKtoBPSK rotator 414 performs phase rotation on each sequence of intermediate samples, rm for m = l, 2, as shown in equation (4), and provides a corresponding sequence of input samples zm. [0043] FIG. 5 shows the two sequences of input samples z\ and zi obtained with 2x oversampling. The early samples in the first sequence z\ are offset by a half sample period from the late samples in the second sequence zi [0044] Referring back to FIG. 4, a cochannel interference suppressor 420 receives the two input sample sequences z\ and z2, suppresses cochannel interference from the undesired transmitter(s), and provides a sequence of CCIsuppressed samples z/. An equalizer 450 performs equalization on the CCIsuppressed samples Zf to suppress intersymbol interference and provides detected bits xaer. Interference suppressor 420 WO 2006/055797 10 PCT/US2005/041856 and equalizer 450 may be implemented in various manners, and several exemplary designs are described below. [0045] Demodulator 160b may perform cochannel interference suppression and equalization for a single iteration or for multiple iterations to improve performance. Each iteration of the cochannel interference suppression and equalization is called an outer iteration. A selector (Sel) 452 receives the training sequence x,s and the detected bits Xdet and provides reference bits xre/ for interference suppressor 420 and equalizer 450. In general, selector 452 may provide the same reference bits to both interference suppressor 420 and equalizer 450 (as shown in FIG. 4) or different reference bits to interference suppressor 420 and equalizer 450 (not shown in FIG. 4). In an embodiment, selector 452 provides the training sequence as the reference bits for the first outer iteration, and provides the training sequence and the detected bits as the reference bits for each subsequent outer iteration. After all of the outer iterations are; completed, RX data processor 170 processes the detected bits for the final outer iteration and generates the decoded data ydec [0046] The received GMSK signal may be assumed to contain the GMSK signal for desired transmitter 110 and an interfering GMSK signal for one undesired transmitter. The input samples from preprocessor 410 may then be expressed as: Eq (6) where a and b represent the input bit sequences at the desired and undesired transmitters, respectively; hm and gm represent the effective channel impulse responses for the desired and undesired transmitters, respectively, for sequence m; and n,„ represents the total noise observed by sequence m. [0047] FIG. 6A shows a model 600 for two transmitters with BPSK. With BPSK, each transmitter transmits realvalued bits instead of complexvalued symbols. For model 600, the realvalued input bits a for the desired transmitter are provided to a channel 610 having a complex channel impulse response h. The realvalued input bits b for the undesired transmitter are provided to a channel 620 having a complex channel impulse response g. The outputs of channels 610 and 620 are added by a summer 630 to generate complexvalued samples z. WO 2006/055797 PCT/US2005/041856 11 [0048] FIG. 6B shows MEMO models 602 and 604 for two transmitters with BPSK. The complex channel impulse response h has a real part /i, and an imaginary part hq. The complex channel impulse response g also has a real part gi and an imaginary part gq. Channel 610 in FIG. 6A is decomposed into a channel 610a having a real channel impulse response hi and a channel 610b having a real channel impulse response hq. Similarly, channel 620 is decomposed into a channel 620a having a real channel impulse response gi and a channel 620b having a real channel impulse response gq. The realvalued input bits a for the desired transmitter are provided to both channels 610a and 610b. The realvalued input bits b for the undesired transmitter are provided to both channels 620a and 620b. A summer 630a sums the outputs of channels 610a and 620a and provides realvalued inphase samples z;. A summer 630b sums the outputs of channels 610b and 620b and provides realvalued quadrature samples zq The inphase samples zi and the quadrature samples zq are the real and imaginary parts, respectively, of the complexvalued samples z. MDVIO model 602 shows a twoinput twooutput (2x2) system being formed with a and b as the two inputs and zi and zq as the two outputs. Two virtual antennas are effectively formed by the real part Zi and the imaginary part zq of z. [0049] The ti and zq samples may be oversampled at multiple (e.g., two) times the sample rate. A demultiplexer 640a demultiplexes the inphase samples zi into two sequences zu and zii, with each sequence containing inphase samples at the sample rate. Similarly, a demultiplexer 640b demultiplexes the quadrature samples zq into two sequences Z\q and Ziq, with each sequence containing quadrature samples at the sample rate. MTMO model 604 shows a twoinput fouroutput (2x4) system being formed with a and b as the two inputs and ZH, Z\q, Zu and Zrq as the four outputs. Four virtual antennas are effectively formed by 2x oversampling the real part zt and the imaginary partz?ofz. [0050] FIG. 7A shows spacetime processing for cochannel interference suppression with MMO model 604 in FIG. 6B. Four virtual antennas are formed with the four realvalued input sample sequences zu, z\q, Zu and ziq obtained with 2x oversampling and real/imaginary decomposition. Using MTMO model 604, appropriate weights may be applied to the four virtual antennas to form a beam toward the direction of the desired transmitter and to create a beam null toward the direction of the undesired transmitter. In general, cochannel interference suppression may be achieved with one WO 2006/055797 PCT/US2005/041856 12 or multiple actual antennas at the receiver by using spacetime processing, where "space" may be virtually achieved with the inphase and quadrature components and "time" may be achieved using late and early samples. [0051] FIG. 7B shows a MIMO filter 700 that performs spacetime processing on two complexvalued input sample sequences Z\ and zi for cochannel interference suppression. A unit 708a receives the complexvalued input sample sequence zu provides the inphase samples zu to FIR filters 710a and 710e, and provides the quadrature samples z\q to FIR filters 710b and 710f, A unit 708b receives the complexvalued input sample sequence Zi, provides the inphase samples zzt to FIR filters 71 (3c and 710g, and provides the quadrature samples ziq to FIR filters 710d and 710h. Each FIR filter 710 has K taps, where K > 1 and may be selected based on the lengths of the channel impulse responses for the desired and undesired transmitters and/or other considerations. [0052] FIG. 7C shows an embodiment of FIR filter 710a within MMO filter 700. FIR filter 710a has K1 seriescoupled delay elements 712b through 712k, K multipliers 714a through 714k, and a summer 716. Each delay element 712 provides one sample period (T) of delay. Multiplier 714a receives the input samples Zu, and multipliers 714b through 714k receive the outputs of delay elements 712b through 712k, respectively. Multipliers 714a through 714k also receive K weights. Each multiplier 714 multiplies its input samples with its weight and provides output samples. For each sample period, summer 716 sums the outputs of all K multipliers 714a through 714k and provides an output sample for that sample period. FIR filters 710b through 710k may each be implemented in the same manner as FIR filter 710a. [0053] Referring back to FIG. 7B, each FIR filter 710 filters its input samples with its set of K real weights w. The weights for FIR filters 710a through 7 lOh are derived to pass the signal from the desired transmitter and to suppress the cochannel interference from the undesired transmitter. A summer 720a sums the outputs of FIR filters 710a through 710d and provides inphase CCIsuppressed samples z/j. A summer 720b sums the outputs of FIR filters 710e through 710h and provides quadrature CCIsuppressed samples z/q. The inphase samples z/? and the quadrature samples z/q may be expressed as: Eq (7a) WO 2006/055797 PCT/US2005/041856 Eq (7b) where wlii, wigq, Wi,;; and W29,/ are four sets of weights for FIR filters 710a, 710b, 710c, and 710d, respectively, and wu,q, wiq,q, wu,q and Wjq.q are four sets of weights for FIR filters 710e, 710f, 710g, and 710h, respectively. Each set contains K weights for the K FIR filter taps. A unit 722 receives the inphase samples Zf, and the quadrature samples z/q and provides complexvalued CCIsuppressed samples zj. [0054] MIMO filter 700 may also be implemented with infinite impulse response (HR) filters or some other type of filter. [0055] In general, multiple virtual antennas may be obtained by (1) oversampling the received signal for each actual antenna to obtain multiple sequences of complexvalued samples and/or (2) decomposing the complexvalued samples into real and imaginary parts. FIG. 6B shows the modeling of two transmitters and a singleantenna receiver as a 2x2 system (with real/imaginary decomposition) and as a 2x4 system (with real/imaginary decomposition and 2x oversampling). For a receiver with Naiu actual antennas, 2NaM virtual antennas may be obtained via real/imaginary decomposition, NaQt • Nos virtual antennas may be obtained via Nos times oversampling, and 2NantNos virtual antennas may be obtained via real/imaginary decomposition and Nos times oversampling. If Nos > 2, then more than two sequences of complexvalued samples may be generated and used to form more than four outputs (and hence more than four virtual antennas) in a MIMO model. For simplicity, the following description is for a receiver with one actual antenna and 2x oversampling. The sequence of received samples rrx is processed to generate four sequences of realvalued input samples Zu, z\q, zu and ziq. [0056] FIG. 8 shows an embodiment of a demodulator 160c that suppresses cochannel interference using virtual antennas. Demodulator 160c may be used for demodulator 160 in FIG. 1. Within demodulator 160c, preprocessor 410 processes the received samples zrx and provides two sequences of complexvalued input samples zi and z%. Demodulator 160c includes a cochannel interference suppressor 420a. and an equalizer 450a. Interference suppressor 420a includes a selector 828, a channel estimator 830, a desired signal estimator 832, a weight computation unit 834, and a WO 2006/055797 PCT/US2005/041856 14 MIMO filter 840. Equalizer 450a includes a channel estimator 860 and a detector 870 (e.g., an MLSE). [0057] Interference suppressor 420a may perform channel estimation and MIMO filtering for a single iteration or for multiple iterations to improve performance. Each iteration of the channel estimation and MIMO filtering is called an inner iteration. Selector 828 receives one sequence of complexvalued input samples (e.g., the first sequence Z\) from preprocessor 410 and the CCIsuppressed sample sequence z/from MIMO filter 840, provides the input sample sequence to channel estimator 830 for the first inner iteration, and provides the CCIsuppressed sample sequence for each subsequent inner iteration. Channel estimator 830 receives the sequence of complexvalued samples (e.g., the first sequence Z\ for the first inner iteration) from selector 828 and the reference bits xref from selector 452 and derives an effective channel impulse response estimate (e.g., h{) for that sequence. Channel estimator 830 may implement a leastsquares (LS) estimator, a linear minimum mean square error (LMMSE), an adaptive filter, or some other type of estimator. In an embodiment that is described below, channel estimator 830 is an LS channel estimator. The input samples for the first sequence z\ may be expressed in vector and matrix form as follows: Eq(8) where h1 =[h0 h1 Ll1 is an Lxl vector with L channel taps for the effective channel impulse response for the desired transmitter for the first sequence z1, where " r" denotes a transpose; X is a Px1 matrix containing the reference bits xref, where P > 1; Z1 is a Pxl vector with P input samples in the first sequence z\, and n1 is a Pxl vector of total noise and interference for the first sequence z1. The effective channel impulse response contains L channel taps hQ through hLl, where L > 1 and each channel tap ht is a complex value. [0058] The reference bits available for channel estimation are arranged into P overlapping segments, with each segment containing L reference bits. The rows of matrix X are formed by the P segments, as follows: WO 2006/055797 15 PCT/US2005/041856 Eq(9) where xnffi through xre/J>_L_2 are PL + l reference bits in xre/. [0059] The LS channel estimator derives a channel impulse response estimate based on the following LS criterion: Eq(10) Equation (10) indicates that hlsd is equal to a hypothesized channel impulse response hi that minimizes the squared error between the input samples zx and the samples generated with that hypothesized channel impulse response (or X • h;). [0060] The solution to equation (10) may be expressed as: WO 2006/055797 PCT/US2005/041856 xref. Signal estimator 832 generates a desired signal estimate si for the desired transmitter by convolving the reference bits with the channel estimate, as follows: Eq(12) The desired signal estimate Si is an estimate of a®hm in equation (6), which is the signal component for the desired transmitter. [0063] Weight computation unit 834 receives the desired signal estimate si and the two input sample sequences Z\ and 22 and derives the weights W\ for MIMO filter 840. MIMO filter 840 may be implemented with MIMO filter 700 having a bank of eight FIR filters 710a through 710h. Unit 834 may compute the weights W\ based on minimum mean square error (MMSE), least squares (LS), or some other criterion. In an embodiment that is described below, unit 834 derives the weights based on the MMSE criterion. [0064] The output of MIMO filter 700 or 840 may be expressed in matrix form as follows: Eq(13) where Z is a 4KxQ matrix of inphase and quadrature samples in sequences z1 zndz2, W is a 2x4K matrix containing the weights for the FDR filters; zf is a 2 xQ matrix of CCIsuppressed samples from the MIMO filter; K is the number of taps for each FIR filter within MIMO filter 700; and Q determines the number of CCIsuppressed samples used to derive the FIR filter weights. Matrices zf, W, and Z may be defined in various manners. An exemplary embodiment for equation (13) is described below. [0065] Matrix Z may be defined with the following form: WO 2006/055797 PCT/US2005/041856 where zu(t) and zlq(t) are respectively the real and imaginary parts of the complexvalued input sample zx(t) in sequence z\ for sample period t; and z2t(t) and z2q(t) are respectively the real and imaginary parts of the complexvalued input sample z2 (0 in sequence z2 for sample period t. Each column of Z contains 4K entries for the real and imaginary parts of 2K complexvalued input samples obtained in K sample periods. Adjacent columns of Z are offset by one sample period. [0066] Matrix W may be defined with the following form: WO 2006/055797 PCT/US2005/041856 18 where zfl(t) and Zfq(0 are the real and imaginary parts of the complexvalued CCI suppressed sample zf (t) for sample period t. The (i, j) th entry of zf is obtained by multiplying the j'th row of W with the ;th column of Z. Each row of zf represents a complexvalued CCIsuppressed sample for one sample period. [0068] Weight computation unit 834 derives the weights for the FIR filters within MIMO filter 840 based on the following MMSE criterion where s is a Wmmse matrix containing Q complexvalued samples in the desired signal estimate S\ provided by signal estimator 832. Equation (17) indicates that Wmme contains the hypothesized weights that minimize the mean squared error between the desired signal estimate s and the CCIsuppressed samples generated with the hypothesized weights (or WZ). [0069] The solution to equation (17) may be expressed as: Hs1 W = sZ (ZZ") Eq(18) The MMSE weights Wm„,w generated based on the desired signal estimate s\ are denoted as Wi. Unit 834 may compute new filter weights for each inner iteration of each outer iteration based on a new desired signal estimate derived for that inner/outer iteration and the two input sample sequences Z\ and^ [0070] MIMO filter 840 receives the two input sample sequences z1 and z2 and the filter weights W\. MTMO filter 840 filters the input samples with the filter weights, as shown in FIG. 7B and equation set (7), and provides the CCIsuppressed samples Zf. MIMO filter 840 suppresses the interference component b ® gm from the undesired transmitter, which results in the CCIsuppressed samples z/ having less cochannel WO 2006/055797 PCT/US2005/041856 interference. However, since the desired signal estimate s\ has intersymbol interference due to the convolution with the channel estimate h{, and since the weights are optimized for the desired signal estimate s\, the CCIsuppressed samples Zf include intersymbol interference. [0071] One or multiple inner iterations may be performed for each outer iteration. For the first inner iteration, the channel estimate h} is derived based on the first sequence z1 and used to generate the filter weights W1 The CCIsuppressed samples z/ are then generated based on the input sample sequences z\ and z2 and the filter weights W\. For each subsequent inner iteration, a new channel estimate h{ is derived based on the CCIsuppressed samples z/ and used to generate new filter weights W1. New CCIsuppressed samples z/ are then generated based on the same input sample sequences zand zz and the new filter weights W\. The new channel estimate hl may have higher quality since it is derived based on the CCIsuppressed samples Zf having cochannel interference suppressed. [0072] Equalizer 450a receives and processes the CCIsuppressed samples Zf from interference suppressor 420a and provides detected bits Xdet. Within equalizer 450a, a channel estimator 860 receives the CCIsuppressed samples zj and the reference bits xref. Equalizer 450a derives an improved estimate of the effective channel impulse response h for the desired transmitter, e.g., based on the LS criterion as shown in equation (10), and provides the improved effective channel impulse response estimate hf to detector 870. Channel estimators 830 and 860 operate in similar manner but on different input sequences. The channel estimate hlf is typically of higher quality than the channel estimate hx because cochannel interference has been suppressed in the sequence Zf used to derive the channel estimate hlf. [0073] Detector 870 performs detection on the CCIsuppressed samples z/with the improved channel estimate hf . Detector 870 may be implemented with an MLSE. In this case, detector 870 convolves hypothesized bits a with the channel estimate hf to generate hypothesized samples z,, which may be expressed as: zf =a®hf . Detector WO 2006/055797 PCT/US2005/041856 870 then computes a branch metric m(i) to be accumulated for each sample period t as follows: Eq(19) zfi(t) and z/q(0 are respectively the real and imaginary parts of the CCI suppressed sample in sequence Z/for sample period t; and Zjj(t) and zfq(t) are respectively the real and imaginary parts of the hypothesized sample in sequence zf for sample period t. Detector 870 provides the detected bits Xdet that are deemed most likely to have been transmitted based on the branch metrics. [0074] The cochannel interference suppression and equalization may be performed once on the input samples zi and zi to obtain the decoded bits ydec Multiple outer iterations of cochannel interference suppression and equalization may also be performed to improve performance. For the first outer iteration, selector 452 provides the training sequence as the reference bits. Channel estimator 830 derives the channel A estimate h1 based on the training sequence. Signal estimator 832 generates the desired signal estimate s\ based on the training sequence and the channel estimate hx. Unit 834 computes the filter weights W1 based on the desired signal estimate si. Channel estimator 860 also derives the improved channel estimate hf based on the training sequence. [0075] For each subsequent outer iteration, selector 452 provides the training sequence and the detected bits as the reference bits. Channel estimator 830 derives the channel estimate /tj based on the training and detected bits. Signal estimator 832 generates a longer desired signal estimate s\ based on the training and detected bits and the channel estimate hx. Unit 834 computes the filter weights W\ based on the longer desired signal estimate. Channel estimator 860 also derives the improved channel estimate hf based on the training and detected bits. After all of the outer iterations are WO 2006/055797 PCT/US2005/041856 21 completed, RX data processor 170 processes the final detected bits Xdet and provides the decoded data ydec [0076] The embodiment in FIG. 8 performs cochannel interference suppression and intersymbol interference suppression separately. This may provide better performance since the MMSEbased MIMO filtering can more effectively deal with cochannel interference while the MLSE can more effectively deal with intersymbol interference. Both types of interference may also be suppressed jointly by providing the reference bits xref (instead of the desired signal estimates s\) to weight computation unit 834. Unit 834 would then compute the weights that minimize the mean square error between the samples from the MIMO filter and the reference bits. [0077] FIG. 9 shows an embodiment of a demodulator 160c that suppresses cochannel interference using virtual antennas and further performs detection with noise decorrelation. For this embodiment, demodulator 160d includes (1) a cochannel interference suppressor 420b that suppresses cochannel interference and provides two sequences of CCIsuppressed samples zi/and Zy and (2) an equalizer 450b that performs data detection on both sequences z\f and zif with noise decorrelation. [0078] Within interference suppressor 420b, a channel estimator 930 receives the two complexvalued input sample sequences z\ and Z2 and the reference bits xref and derives effective channel impulse response estimates /i, and h2 for sequences Z\ and z2, respectively. Each channel estimate hm, for m = l, 2, may be derived based on the input sample sequence zm and using the LS criterion, as shown in equation (10). A desired signal estimator 932 receives the reference bits xre/and the channel estimates hl and h2, derives a desired signal estimate s1 based on xre/and hx as shown in equation (12), derives a desired signal estimate s2 based on xref and h2, and provides the two desired signal estimates s1 and s%. [0079] . A weight computation unit 934 receives the input sample sequences z\ and zi and the desired signal estimates s\ and s2 and generates weights \V\ and W2 for a MIMO filter 940. MIMO filter 940 may be implemented with two instances of MIMO filter 700 shown in FIG. 7B, which are called first and second MMO filters. The first MIMO filter filters the input sample sequences Z\ and z2 with the weights W\, as shown in equation set (7), and provides a first CCIsuppressed sample sequence Z1f. The WO 2006/055797 PCT/US2005/041856 second MIMO filter filters the input sample sequences z\ and z% with the weights W2 and provides a second CCIsuppressed sample sequence zif. The first and second MIMO filters operate independently of one another. Unit 934 derives the weights W\ such that the mean square error between the CCIsuppressed samples zi/and the desired signal estimate s\ is minimized, as shown in equation (17). Unit 934 derives the weights W2 such that the mean square error between the CCIsuppressed samples zy and the desired signal estimate S2 is minimized. [0080] For clarity, FIG. 9 shows interference suppressor 420b performing one inner iteration of channel estimation and MIMO filtering. Interference suppressor 420b may also perform multiple inner iterations to improve performance. In this case, a selector can receive the two input sample sequences z\ and Z2 from preprocessor 410 and the two CCIsuppressed sample sequences Z\j and Z2/ from MEMO filter 940, provide the input sample sequences Z\ and zi to channel estimator 930 for the first inner iteration, and provide the CCIsuppressed sample sequences zy and Z2/for each subsequent inner iteration. [0081] Within equalizer 450b, a channel estimator 960 receives the two CCIsuppressed sample sequences z\j and Z2/ and the reference bits xref and derives improved A A effective channel impulse response estimates hlf and h2f for sequences z\f and Z2/, respectively. Each channel estimate hmf, for m = 1, 2, may be derived based on CCIsuppressed sample sequence zm/ and using the LS criterion, as shown in equation (10). The channel estimates hlf and h2f are typically of higher quality than the channel A A estimates /i, and h2 because cochannel interference has been suppressed in the A A sequences zy and Z2/used to derive the channel estimates hlf and h2f . [0082] A desired signal estimator 962 receives the reference bits xref and the A A improved channel estimates hlf and h2f , derives a desired signal estimate si/based on x and hlf as shown in equation (12), derives a desired signal estimate s2f based on xref and h2f , and provides the two desired signal estimates Sy and S1y. Signal estimators 932 and 962 operate in similar manner but with different channel estimates. The desired signal estimates 1y and sy are typically of higher quality than the desired signal WO 2006/055797 PCT/US2005/041856 estimates S\ and S2 because of the improved channel estimates hlf and h2f used to derive the desired signal estimates s 1/ and s2f. [0083] A summer 964a subtracts the desired signal estimate s\f from the CCIsuppressed samples z\f and provides a noise estimate ii\p A summer 964b subtracts the desired signal estimate s2/ from the CCIsuppressed samples zy and provides a noise estimate mp The noise estimates may be expressed as: [0084] A computation unit 966 computes a 4x4 noise correlation matrix R„„ based on the .real and imaginary parts of the noise samples in lyand ny, as follows: R„„=(n,nr) , Eq(21) where n, ~[nlfl{t) nlfq(t) n2fl{t) n2fq(t)]T is a 4x1 noise vector for sample period t; nXfi(t) and nlfq(t) are the real and imaginary parts of the noise sample in /ii/for sample period t; n2fi{t) and n2fq(t) are the real and imaginary parts of the noise sample in ny for sample period t; and ( ) denotes an averaging operation. [0085] A detector 970 receives the CCIsuppressed sample sequences Zi/and zip the improved channel estimates hlf and h2f, and the noise correlation matrix R„„. Detector 970 performs detection based on all of the inputs. Detector 970 may be implemented with an MLSE. In this case, detector 970 convolves hypothesized bits a with the channel estimate hlf to derive a first sequence of hypothesized samples zif (or zlf =a®hlf). Detector 970 also convolves the hypothesized bits a with the channel estimate h2f to derive a second sequence of hypothesized samples z2f (or z2f = a ® h2f). Detector 970 then computes the branch metric m(t) to be accumulated for each sample period t as follows: WO 2006/055797 PCT/US2005/041856 ziyr(t) and z1/9(0 are respectively the real and imaginary parts of the CCI suppressed sample in sequence zi/for sample period t; z2fi(t) and zlfq(t) are respectively the real and imaginary parts of the CCI suppressed sample in sequence Z2/for sample period /; zXfl(t) and zlf(t) are respectively the real and imaginary parts of the hypothesized sample in sequence zlf for sample period t; and z2fi(t) and z2fqa(t) are respectively the real and imaginary parts of the hypothesized sample in sequence z2f for sample period t. Equation (22) incorporates spatial decorrelation into the branch metrics used by the MLSE. Detector 970 provides the detected bits Xdet that are deemed most likely to have ' been transmitted based on the branch metrics. [0086] For the embodiments shown in FIGS. 8 and 9, the same reference bits xref are provided to both the cochannel interference suppressor and the equalizer and are used to derive the channel estimates and the desired signal estimates. In general, the same or different reference bits may be provided to the cochannel interference suppressor and the equalizer. Furthermore, the same or different reference bits may be used for channel estimation and desired signal estimation. [0087] For the embodiment shown in FIG. 8, a new channel estimate and a new desired signal estimate are derived for each inner iteration of each outer iteration. For the embodiment shown in FIG. 9, a new channel estimate and a new desired signal estimate are derived for each outer iteration. In general, new or prior channel estimates may be used for each inner and outer iteration, and new or prior desired signal estimates may be used for each inner and outer iteration. For example, the channel estimates hx and h2 may be derived once based on the training sequence and used for all outer iterations. WO 2006/055797 PCT/US2005/041856 [0088] For the embodiments shown in FIGS. 4, 8 and 9, the detected bits xdet from the equalizer are used to derive the channel estimates (e.g., /i,, h2, hif and h2f in FIG. 9) and the desired signal estimates (e.g., S[, sj, S\j and «2/ in FIG. 9) for a subsequent outer iteration. Some of the detected bits may be of low quality and would then degrade the quality of the channel estimates and the desired signal estimates. Improved performance may be achieved by identifying detected bits of low quality and selectively discarding these bits. [0089] FIG. 10 shows an embodiment of a demodulator 160e that performs interference suppression using bit pruning. Demodulator 160e includes all of the elements of demodulator 160b in FIG. 4. However, demodulator 160e utilizes a different feedback mechanism for the reference bits. [0090] Within demodulator 160e, a filter 1080 receives the soft decisions from soft output generator 380 and a channel estimate (e.g., hf) from equalizer 450. Each soft decision indicates the confidence in a corresponding detected bit. Filter 1080 may be implemented with an Ltap FIR filter having a length corresponding to the length of the channel estimate. In an embodiment, the weights q for the L taps of the FIR filter are derived based on the L taps of the channel estimate, as follows: qk is the weight for the /cth tap of the FIR filter. With the weights generated in accordance with equation (23), filter 1080 implements a channel energy filter having normalized filter taps so that [0091] Filter 1080 filters the magnitude of the soft decisions with its weights q and provides filtered symbols. A threshold compare unit 1082 compares each filtered symbol against a threshold value and indicates whether the filtered symbol is greater than the threshold value. Because of the normalization in equation (23), the threshold WO 2006/055797 PCT/US2005/041856 value may be set to a predetermined value (e.g., 10 decibel) that is independent of the actual taps for the channel estimate. The threshold value may be determined by computer simulation, empirical measurements, and so on. [0092] A pruning unit 1084 receives the indications from threshold compare unit 1082 and the detected bits Xde, from equalizer 450 and provides unpruned bits x,h, which may be used as the reference bits for channel estimation and desired signal estimation. Unit 1084 generates the unpruned bits in a manner to account for the processing performed by interference suppressor 420 and equalizer 450. As an example, for each filtered symbol that is deemed to be of poor quality, a column of matrix X corresponding to that filtered symbol may be deleted (or set to all zeros) and not used for channel estimation. The overall effect of bit pruning is to use detected bits having good quality for cochannel interference suppression and equalization and to remove (or prune) detected bits with poor quality from being used. The channel energy filter removes poor quality detected bits only when these bits have a relatively large impact, e.g., when these bits are multiplied with a large channel tap. Selector 452 receives the training bits xts and the unpruned bits x,iu provides the training bits as the reference bits Xref for the first outer iteration, and provides the training bits and the unpruned bits as the reference bits for each subsequent outer iteration. [0093] FIG. 10 shows a specific embodiment for determining the quality of the equalizer output and for generating the reference bits based on the determined quality. The quality of the equalizer output may also be determined in other manners using other detection schemes. The reference bits may also be generated in other manners. [What are some other ways to determine quality? What are some other ways to generate the reference bits?] [0094] For the embodiments shown in FIGS. 4, 8, 9 and 10, the unpruned or pruned detected bits Xdet from the equalizer are used in each subsequent outer iteration to derive the channel,estimates and the desired signal estimates. Improved performance may be achieved by using the error correction capability of the forward error correction (FEC) code to feed back higher quality bits for cochannel interference suppression and equalization. [0095] FIG. 11 shows an embodiment of a demodulator 160f that performs interference suppression using reencoded bits. Demodulator 160f includes all of the WO 2006/055797 PCT/US200S/04856 elements of demodulator 160b in FIG. 4. However, demodulator 160e utilizes a different feedback mechanism that uses reencoded bits. [0096] For each outer iteration except for the last outer iteration, RX data processor 170 processes the detected bits x^, from demodulator 160f and provides decoded bits ydec TX data processor 120 reencodes and interleaves the decoded bits ydec in the same manner performed by transmitter 110 and generates reencoded bits xenc The reencoded bits are typically of higher quality than the detected bits Xdet because the Viterbi decoder within RX data processor 170 is typically able to correct some or many of the bit errors. Selector 452 receives the training bits xts and the reencoded bits xenc, provides the training bits as the reference bits xref for the first outer iteration, and provides the training bits and the reencoded bits as the reference bits for each subsequent outer iteration. [0097] FIG. 12 shows an embodiment of a demodulator I60g and an RX data processor 170b that perform iterative interference suppression and decoding. Within demodulator 160g, a cochannel interference suppressor 1220 receives the two complexvalued input sample sequences z\ and zi from preprocessor 410 and possibly soft outputs yS0i from an interleaver 1286. Interference suppressor 1220 may be implemented with interference suppressor 420a in FIG. 8, interference suppressor 420b in FIG. 9, or some other design. Interference suppressor 1220 suppresses cochannel interference and provides CCIsuppressed samples. A softoutput equalizer 1250 performs equalization on the CCIsuppressed samples and possibly the soft outputs ysoi and provides soft detected symbols xsu Interference suppressor 1220 and equalizer 1250 may use the soft outputs ySOi in various manners. For example, the soft outputs ySOi may be used for channel estimation. As another example, equalizer 1250 may implement a softinput softoutput equalizer that utilizes the information in the soft outputs ysoi to improve detection performance. [0098] Within RX data processor 170b, a deinterleaver 1282 deinterleaves the soft detected symbols xso in a manner complementary to the interleaving performed by the desired transmitter 110. A soft output Viterbi algorithm (SOVA) decoder 1284 performs decoding on the deinterleaved symbols from deinterleaver 1282, provides soft outputs yso for each outer iteration except for the last outer iteration, and provides decoded bits ydec for the last outer iteration. Interleaver 1286 interleaves the soft outputs WO 2006/055797 PCT/US200STO41856 yso from SOVA decoder 1284 in the same manner performed by TX data processor 120 at transmitter 110 and provides the interleaved soft outputs ysoi. [0099] For the embodiment shown in FIG. 12, interference suppressor 1220 and soft output equalizer 1250 form, a softinput softoutput (SISO) detector 1210. SISO detector 1210 receives soft inputs from preprocessor 410 and soft inputs from SOVA decoder 1284 via interleaver 1286, suppresses cochannel interference and intersymbol interference, and provides soft outputs. This embodiment performs iterative interference suppression (via SISO detector 1210) and decoding (via SOVA decoder 1284) to achieve improved performance. This structure also resembles a Turbo decoder with two SISO decoders coupled in a feedback configuration. [00100] For clarity, specific embodiments of a receiver with a single actual antenna have been described above for GSM. In general, the receiver may be equipped with any number of actual antennas that may be used to form any number of virtual antennas. The receiver may also be used for various communication systems such as a Time Division Multiple Access (TDMA) system, a Code Division Multiple Access (CDMA) system, a Frequency Division Multiple Access (FDMA) system, an Orthogonal Frequency Division Multiple Access (OFDMA) system, and so on. A TDMA system may implement one or more TDMA radio access technologies (RATs) such as GSM. A CDMA system may implement, one or more CDMA RATs such as WidebandCDMA (WCDMA), cdma2000, and TSCDMA. These various RATs are well known in the art. WCDMA and GSM are parts of Universal Mobile Telecommunication System (UMTS) and are described in documents from a consortium named "3rd Generation Partnership Project" (3GPP). cdma2000 is described in documents from a consortium named "3rd Generation Partnership Project 2" (3GPP2). 3GPP and 3GPP2 documents are publicly available. The innovative receiver provides improved performance over conventional receivers and allows a network to improve capacity by using the same frequency band/channel at shorter distances. [00101] The receiver described herein may be implemented by various means. For example, the receiver may be implemented in hardware, software, or a combination thereof. For a hardware implementation, the processing units used to perform cochannel interference suppression, equalization, and data processing may be implemented within one or more application specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable WO 2006/055797 PCT/US2005/041856 logic devices (PLDs), field programmable gate arrays (FPGAs), processors, controllers, microcontrollers, microprocessors, other electronic units designed to perform the functions described herein, or a combination thereof. [00102] For a software implementation, the processing may be implemented with modules (e.g., procedures, functions, and so on) that: perform the functions described herein. The software codes may be stored in a memory unit (e.g., memory unit 182 in FIG. 1) and executed by a processor (e.g., controller 180). The memory unit may be implemented within the processor or external to the processor. [00103] The previous description of the disclosed embodiments is provided to enable any person skilled in the art to make or use the present invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other embodiments without departing from the spirit or scope of the invention. Thus, the present invention is not intended to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein. [00104] WHAT IS CLAIMED IS: WO 2006/055797 PCT/US2005/041856 30 We Claim 1. A receiver comprising: a preprocessor operative to process received samples for at least one antenna to generate a plurality of sequences of input samples, wherein at least two sequences of input samples are generated for each antenna; an interference suppressor operative to suppress cochannel interference (CCI) in the plurality of sequences of input samples and to provide at least one sequence of CCIsuppressed samples; and an equalizer operative to perform detection on the at least one sequence of CCIsuppressed samples. 2. The receiver of claim 1, wherein the preprocessor is operative to provide at least one sequence of early samples and at least one sequence of late samples for each antenna, the early and late samples being offset by a half sample period. 3. The receiver of claim 1, wherein the preprocessor is operative to provide at least one sequence of inphase samples and at least one sequence of quadrature samples for each antenna, the inphase and quadrature samples corresponding to real and imaginary parts, respectively, of complexvalued samples. 4. The receiver of claim 1, wherein the preprocessor is operative to process received samples for a single antenna to generate at least two sequences of input samples for the single antenna. 5. , The receiver of claim 1, wherein the preprocessor comprises a filter operative to filter the received samples to generate a plurality of sequences of intermediate samples, and a rotator operative to perform phase rotation on the plurality of sequences of intermediate samples to generate the plurality of sequences of input samples. WO 2006/055797 PCT/US2005/041856 6. The receiver of claim 1, wherein the interference suppressor comprises a multipleinput multipleoutput (MLMO) filter operative to filter the plurality of sequences of input samples with a plurality of weights to generate the at least one sequence of CCIsuppressed samples, the plurality of weights being derived to suppress the cochannel interference in the plurality of sequences of input samples. 7. The receiver of claim 6, wherein the MTMO filter comprises a plurality of finite impulse response (FIR) filters operative to filter the plurality of sequences of input samples. 8. The receiver of claim 7, wherein each of the plurality of FIR filters is operative to filter a respective sequence of input samples with a respective set of weights. 9. The receiver of claim 1, wherein the interference suppressor comprises a channel estimator operative to derive at least one channel estimate based on the plurality of sequences of input samples, a signal estimator operative to derive at least one desired signal estimate based on the at least one channel estimate, a computation unit operative to compute weights used to suppress cochannel interference, and a multipleinput multipleoutput (MMO) filter operative to filter the plurality of sequences of input samples with the weights to generate the at least one sequence of CCIsuppressed samples. 10. The receiver of claim 9, wherein the channel estimator is operative to derive the at least one channel estimate using a leastsquares (LS) criterion. 11. The receiver of claim 9, wherein the computation unit is operative to compute the weights for the MTMO filter using a minimum mean square error (MMSE) criterion. WO 2006/055797 PCT/US2005/041856 32. 12. The receiver of claim 9, wherein the channel estimator, the signal estimator, the computation unit, and the MIMO filter are operated for a plurality of iterations. 13. The receiver of claim 1, wherein the equalizer comprises a channel estimator operative to derive at least one channel estimate based on the at least one sequence of CCIsuppressed samples, and a detector operative to perform detection on the at least one sequence of CCIsuppressed samples with the at least one channel estimate. 14. The receiver of claim 13, wherein the detector is a maximum likelihood sequence estimator (MLSE). 15. The receiver of claim 1, wherein the interference suppressor is operative to provide at least two sequences of CCIsuppressed samples, and wherein the equalizer comprises a channel estimator operative to derive at least two channel estimates based on the at least two sequences of CCIsuppressed samples, a signal estimator operative to derive at least two desired signal estimates based on the at least two channel estimates, s a computation unit operative to compute a noise correlation matrix based on the at least two desired signal estimates and the at least two sequences of CCIsuppressed samples, and a detector operative to perform detection on the at least two sequences of CCIsuppressed samples with the at least two channel estimates and the noise correlation matrix. 16. The receiver of claim 15, wherein the detector is a maximum likelihood sequence estimator (MLSE) and is operative to compute branch metrics using the noise correlation matrix. WO 2006/055797 PCT/US2005/041856 33 17. The receiver of claim 1, further comprising: a selector operative to receive detected bits from the equalizer and a training sequence and to provide reference bits for the interference suppressor and the equalizer. 18. The receiver of claim 1, wherein the interference suppressor and the equalizer are operated for a plurality of iterations. 19. The receiver of claim 18, wherein the interference suppressor is operative to suppress the cochannel interference based on a training sequence for a first iteration and based on detected bits from the equalizer and the training sequence for each subsequent iteration. 20. The receiver of claim 18, wherein the equalizer is operative to perform detection based on a training sequence for a first iteration and based on detected bits from the equalizer and the training sequence for each subsequent iteration. 21. The receiver of claim 1, further comprising: a filter operative to filter soft decisions generated based on an output of the equalizer and to provide filtered symbols; a threshold compare unit operative to compare the filtered symbols against a threshold and to provide comparison results; and a selector operative to provide reference bits for the interference suppressor and the equalizer based on the comparison results. 22. The receiver of claim 21, wherein the filter is operative to filter the soft decisions with a plurality of weights derived based on a channel impulse response estimate. 23. The receiver of claim 1, further comprising: a receive data processor operative to process an output of the equalizer to obtain decoded data; and WO 2006/055797 PCT/US2005/041856 34 a transmit data processor operative to process the decoded data to generate reencoded bits, wherein the equalizer is operative to perform detection based on the reencoded bits. 24. The receiver of claim 23, wherein the interference suppressor is operative to suppress cochannel interference based on the encoded bits. 25. The receiver of claim 1, further comprising: a receive data processor operative to process an output of the equalizer to generate soft output symbols for a softinput softoutput (SISO) detector formed by the interference suppressor and the equalizer, wherein the SISO detector and the receive data processor are operated for a plurality of iterations. 26. A receiver comprising: a preprocessor operative to process received samples for at least one antenna to generate a plurality of sequences of input samples, wherein at least two sequences of input samples are generated for each antenna by oversampling a received signal for the antenna, decomposing complexvalued samples into inphase and quadrature samples, or both oversampling and decomposing; an interference suppressor operative to suppress cochannel interference (CCI) in the plurality of sequences of input samples and to provide at least one sequence of CCIsuppressed samples; and an equalizer operative to perform detection on the at least one sequence of CCIsuppressed samples. 27. A method of receiving data in a communication system, comprising: processing received samples for at least one antenna to generate a plurality of sequences of input samples, wherein at least two sequences of input samples are generated for each antenna; suppressing cochannel interference (CCI) in the plurality of sequences of input samples to generate at least one sequence of CCIsuppressed samples; and performing detection on the at least one sequence of CCIsuppressed samples. WO 2006/055797 PCT/US2005/041856 28. The method of claim 27, wherein the processing the received samples for the at least one antenna comprises filtering the received samples to obtain a plurality of sequences of intermediate samples, and performing phase rotation on the plurality of sequences of intermediate samples to generate the plurality of sequences of input samples. 29. The method of claim 27, wherein the suppressing cochannel interference in the plurality of sequences of input samples comprises computing a plurality of weights used to suppress cochannel interference, and filtering the plurality of sequences of input samples with the plurality of weights to generate the at least one sequence of CCIsuppressed samples. 30. The method of claim 27, wherein the suppressing cochannel interference in the plurality of sequences of input samples comprises deriving at least one channel estimate based on the plurality of sequences of input samples, deriving at least one desired signal estimate based on the at least one channel estimate, computing weights used to suppress cochannel interference, and filtering the plurality of sequences of input samples with the weights to generate the at least one sequence of CCIsuppressed samples. 31. The method of claim 27, wherein the performing detection on the at least one sequence of CCIsuppressed samples comprises deriving a channel estimate based on the at least one sequence of CCIsuppressed samples, and performing detection on the at least one sequence of CCIsuppressed samples with the channel estimate. 32. The method of claim 27, wherein at least two sequences of CCI suppressed samples are generated, and wherein the performing detection on the at least two sequences of CCIsuppressed samples comprises w0 2006/055797 PCT/US2005/041856 deriving at least two channel estimates based on the at least two sequences of CCIsuppressed samples, deriving at least two desired signal estimates based on the at least two channel estimates, computing a noise correlation matrix based on the at least two desired signal estimates and the at least two sequences of CCIsuppressed samples, and performing detection on the at least two sequences of CCIsuppressed samples with the at least two channel estimates and the noise conelation matrix. 33. The method of claim 27, further comprising: performing cochannel interference suppression and detection for a plurality of iterations. 34. The method of claim 27, further comprising: determining quality of detected bits generated by the detection; generating reference bits based on the determined quality of the detected bits; and using the reference bits for cochannel interference suppression, detection, or both cochannel interference suppression and detection. 35. The method of claim 27, further comprising: decoding detected bits generated by the detection to obtain decoded data; encoding the decoded data to obtain reencoded bits; and using the reencoded bits for cochannel interference suppression, detection, or both cochannel interference suppression and detection. 36. The method of claim 27, further comprising: decoding an output generated by the detection to obtain soft output symbols; and using the soft output symbols for cochannel interference suppression, detection, or both cochannel interference suppression and detection. WO 2006/055797 PCT/US2005/041856 37. An apparatus in a communication system, comprising: means for processing received samples for at least one antenna to generate a plurality of sequences of input samples, wherein at least two sequences of input samples are generated for each antenna; means for suppressing cochannel interference (CCI) in the plurality of sequences of input samples to generate at least one sequence of CCIsuppressed samples; and means for performing detection on the at least one sequence of CCIsuppressed samples. 38. The apparatus of claim 37, wherein the means for processing the received samples for the at least one antenna comprises means for filtering the received samples to obtain a plurality of sequences of intermediate samples, and means for performing phase rotation on the plurality of sequences of intermediate samples to generate the plurality of sequences of input samples. 39. The apparatus of claim 37, wherein the means for suppressing co channel interference in the plurality of sequences of input samples comprises means for computing a plurality of weights used to suppress cochannel interference, and means for filtering the plurality of sequences of input samples with the plurality of weights to generate the at least one sequence of CCIsuppressed samples. 40. The apparatus of claim 37, wherein the means for suppressing co^ channel interference in the plurality of sequences of input samples comprises means for deriving at least one channel estimate based on the plurality of sequences of input samples, means for deriving at least one desired signal estimate based on the at least one channel estimate, means for computing weights used to suppress cochannel interference, and WO 2006/055797 PCT/US2005/041856 means for filtering the plurality of sequences of input samples with the weights to generate the at least one sequence of CCIsuppressed samples. 41. The apparatus of claim 37, wherein the means for performing detection on the at least one sequence of CCIsuppressed samples comprises means for deriving a channel estimate based on the at least one sequence of CCIsuppressed samples, and means for performing detection on the at least one sequence of CCIsuppressed samples with the channel estimate. 42. The apparatus of claim 37, wherein at least two sequences of CCI suppressed samples are generated, and wherein the means for performing detection on the at least two sequences of CCIsuppressed samples comprises means for deriving at least two channel estimates based on the at least two sequences of CCIsuppressed samples, means for deriving at least two desired signal estimates based on the at least two channel estimates, means for computing a noise correlation matrix based on the at least two desired signal estimates and the at least two sequences of CCIsuppressed samples, and means for performing detection on the at least two sequences of CCIsuppressed samples with the at least two channel estimates and the noise correlation matrix. 43. The apparatus of claim 37, further comprising: means for performing cochannel interference suppression and detection for a plurality of iterations. 44. The apparatus of claim 37, further comprising: means for determining quality of detected bits generated by the detection; means for generating reference bits based on the determined quality of the detected bits; and means for using the reference bits for cochannel interference suppression, detection, or both cochannel interference suppression and detection. WO 2006/055797 PCT/US2005/041856 45. The apparatus of claim 37, further comprising: means for decoding detected bits generated by the detection to obtain decoded data; means for encoding the decoded data to obtain reencoded bits; and means for using the reencoded bits for cochannel interference suppression, detection, or both cochannel interference suppression and detection. 46. The apparatus of claim 37, further comprising: means for decoding an output generated by the detection to obtain soft output symbols; and  means for using the soft output symbols for cochannel interference suppression, detection, or both cochannel interference suppression and detection. Dated this 23rd day of May, 2007. ABSTRACT INTERFERENCE SUPPRESSION WITH VIRTUAL ANTENNAS A receiver suppresses cochannel interference (CCI) from other transmitters and intersymbol interference (ISI) due to channel distortion using 'virtual' antennas. The virtual antennas may be formed by (1) oversampling a received signal for each actual antenna at the receiver and/or (1) decomposing a sequence of complexvalued samples into a sequence of inphase samples and a sequence of quadrature samples. In one design, the receiver includes a preprocessor, an interference suppressor, and an equalizer. The preprocessor processes received samples for at least one actual antenna and generates at least two sequences of input samples for each actual antenna. The interference suppressor suppresses cochannel interference in the input sample sequences and provides at least one sequence of CCIsuppressed samples. The equalizer performs detection on the CCIsuppressed sample sequence(s) and provides detected bits. The interference suppressor and equalizer may be operated for one or multiple iterations. 

764mumnp2007abstract(granted)(2522010).pdf
764MUMNP2007CANCELLED PAGES(17112009).pdf
764mumnp2007cancelled pages(2422010).pdf
764MUMNP2007CLAIMS(AMENDED)(17112009).pdf
764MUMNP2007CLAIMS(AMENDED)(2422010).pdf
764mumnp2007claims(granted)(2522010).pdf
764mumnp2007correspondence(16112007).pdf
764MUMNP2007CORRESPONDENCE(23112009).pdf
764MUMNP2007CORRESPONDENCE(24112009).pdf
764mumnp2007correspondence(ipo)(2522010).pdf
764mumnp2007correspondencereceived.pdf
764mumnp2007description (complete).pdf
764mumnp2007description(granted)(2522010).pdf
764MUMNP2007DRAWING(17112009).pdf
764mumnp2007drawing(2552007).pdf
764mumnp2007drawing(granted)(2522010).pdf
764MUMNP2007FORM 1(17112009).pdf
764mumnp2007form 1(2552007).pdf
764mumnp2007form 13(17112009).pdf
764mumnp2007form 13(copy)(17112009).pdf
764MUMNP2007FORM 16(8122011).pdf
764mumnp2007form 2(granted)(2522010).pdf
764mumnp2007form 2(title page)(2552007).pdf
764mumnp2007form 2(title page)(granted)(2522010).pdf
764mumnp2007form 3(16112007).pdf
764MUMNP2007FORM 3(17112009).pdf
764MUMNP2007FORM 3(24112009).pdf
764mumnp2007form 3(2552007).pdf
764MUMNP2007OTHER DOCUMENT(17112009).pdf
764MUMNP2007PETITION UNDER RULE 137(23112009).pdf
764MUMNP2007REPLY TO EXAMINATION REPORT(17112009).pdf
764MUMNP2007REPLY TO HEARING(2422010).pdf
764mumnp2007wo international publication report(2552007).pdf
Patent Number  238903  

Indian Patent Application Number  764/MUMNP/2007  
PG Journal Number  10/2010  
Publication Date  05Mar2010  
Grant Date  25Feb2010  
Date of Filing  25May2007  
Name of Patentee  QUALCOMM INCORPORATED  
Applicant Address  5775 MOREHOUSE DRIVE SAN DIEGO, CALIFORNIA 921211714  
Inventors:


PCT International Classification Number  H04L25/03  
PCT International Application Number  PCT/US2005/041856  
PCT International Filing date  20051117  
PCT Conventions:
