Title of Invention	AN APPARATUS FOR PERFORMING CHANNEL AND NOISE ESTIMATION FOR A MULTIPLE -INPUT MULTIPLE -OUTPUT(MIMO) TRANSMISSION
Abstract	Techniques for performing channel and noise estimation for a MIMO transmission sent from multiple transmit antennas to multiple receive antennas are described. Samples are obtained from the receive antennas. For a first scheme, channel estimates are derived by correlating the samples with at least one pilot sequence, and signal, noise and interference statistics are also estimated based on the samples. For a second scheme, total received energy as well as signal and interference energy are estimated based on the samples. Noise is then estimated based on the estimated total received energy and the estimated signal and interference energy. For a third scheme, signal and on-time interference statistics are estimated based on the samples. Noise and multipath interference statistics are also estimated based on the samples. Signal, noise and interference statistics are then estimated based on the estimated signal and on-time interference statistics and the estimated noise and multipath interference statistics.

Title of Invention

AN APPARATUS FOR PERFORMING CHANNEL AND NOISE ESTIMATION FOR A MULTIPLE -INPUT MULTIPLE -OUTPUT(MIMO) TRANSMISSION

Abstract

Techniques for performing channel and noise estimation for a MIMO transmission sent from multiple transmit antennas to multiple receive antennas are described. Samples are obtained from the receive antennas. For a first scheme, channel estimates are derived by correlating the samples with at least one pilot sequence, and signal, noise and interference statistics are also estimated based on the samples. For a second scheme, total received energy as well as signal and interference energy are estimated based on the samples. Noise is then estimated based on the estimated total received energy and the estimated signal and interference energy. For a third scheme, signal and on-time interference statistics are estimated based on the samples. Noise and multipath interference statistics are also estimated based on the samples. Signal, noise and interference statistics are then estimated based on the estimated signal and on-time interference statistics and the estimated noise and multipath interference statistics.

Full Text	METHOD AND APPARATUS FOR CHANNEL AND NOISE ESTIMATION 1. Claim of Priority under 35 U.S.C. §119 [0001] The present Application for Patent claims priority to Provisional Application Serial No. 60/731,423, entitled "METHOD AND APPARATUS FOR SPACE-TIME EQUALIZATION IN WIRELESS COMMUNICATIONS," filed October 28, 2005, assigned to the assignee hereof, and expressly incorporated herein by reference. BACKGROUND I. Field [0002] The present disclosure relates generally to communication, and more specifically to techniques for performing channel and noise estimation for a multiple-input multiple-output (MEMO) transmission. II. Background [0003] A MIMO transmission is a transmission from multiple (T) transmit antennas to multiple (R) receive antennas. For example, a transmitter may simultaneously transmit T data streams from the T transmit antennas. These data streams are distorted by the propagation environment and further degraded by noise. A receiver receives the transmitted data streams via the R receive antennas. The received signal from each receive antenna contains scaled versions of the transmitted data streams, possibly at different delays determined by the propagation environment. The transmitted data streams are thus dispersed among the R received signals from the R receive antennas. The receiver would then perform receiver spatial processing (e.g., space-time equalization) on the R received signals in order to recover the transmitted data streams. [0004] The receiver may derive channel and noise estimates for the MIMO channel and may then derive weights for a space-time equalizer based on the channel and noise estimates. The quality of the channel and noise estimates may have a large impact on performance. There is therefore a need in the art for techniques to derive quality channel and noise estimates for a MIMO transmission. SUMMARY [0005] According to an embodiment of the invention, an apparatus is described which includes at least one processor and a memory. The processors) obtain samples from multiple receive antennas for a MIMO transmission sent from multiple transmit antennas, derive channel estimates by correlating the samples with at least one pilot sequence, and estimate signal, noise and interference statistics based on the samples. The MIMO transmission may comprise multiple modulated signals sent from the multiple transmit antennas. Each modulated signal may comprise multiple data streams multiplexed with different orthogonal codes. [0006] According to another embodiment, an apparatus is described which includes at least one processor and a memory. The processor's) obtain samples from multiple receive antennas for a MIMO transmission sent from multiple transmit antennas, estimate total received energy based on the samples, estimate signal and interference energy based on the samples, and estimate noise based on the estimated total received energy and the estimated signal and interference energy. [0007] According to yet another embodiment, an apparatus is described which includes at least one processor and a memory. The processor's) obtain samples from multiple receive antennas for a MIMO transmission sent from multiple transmit antennas, estimate signal and on-time interference statistics based on the samples, estimate noise and multipath interference statistics based on the samples, and estimate signal, noise and interference statistics based on the estimated signal and on-time interference statistics and the estimated noise and multipath interference statistics. The on-time interference is interference that arrives at the same time as a desired signal, and the multipath interference is interference that is not on time. [0008] According to yet another embodiment, an apparatus is described which includes at least one processor and a memory. The processor's) obtain samples from multiple receive antennas for a MIMO transmission sent from multiple transmit antennas, determine the channel conditions based on the samples, select one of multiple channel and noise estimation schemes based on the channel conditions, and perform channel and noise estimation based on the selected channel and noise estimation scheme. To determine the channel conditions, the processors) may process the samples to determine delay spread, to identify single path or multipath environment, and so on. [0009] Various aspects and embodiments of the invention are described in further detail below. BRIEF DESCRIPTION OF THE DRAWINGS [0010] FIG. 1 shows a transmitter and a receiver for a MIMO transmission. [0011] FIG. 2 shows a CDMA modulator for one transmit antenna. [0012] FIG. 3 shows a channel and noise estimator for a first scheme. 10013J FIG. 4 shows a channel estimator. [0014] F7G. 5 shows a process for the first channel and noise estimation scheme. [0015] FIG. 6 shows a channel and noise estimator for a second scheme. [0016] FIG. 7 shows a process for the second channel and noise estimation scheme. [0017] FIG. 8 shows a channel and noise estimator for a third scheme. [0018] FIG. 9 shows a process for the third channel and noise estimation scheme. [0019] FIG. 10 shows a process for a fourth channel and noise estimation scheme. DETAILED DESCRIPTION [0020] The word "exemplary" is used herein to mean "serving as an example, instance, or illustration." Any embodiment described herein as "exemplary" is not necessarily to be construed as preferred or advantageous over other embodiments. [0021] The channel and noise estimation techniques described herein may be used for various communication systems such as Code Division Multiple Access (CDMA) systems, Time Division Multiple Access (TDMA) systems, Frequency Division Multiple Access (FDMA) systems, Orthogonal FDMA (OFDMA) systems, Single-Carrier FDMA (SC-FDMA) systems, and so on. A CDMA system may implement one or more radio technologies such as Wideband-CDMA (W-CDMA), cdma2000, and so on. cdma2000 covers 1S-2000, IS-856, and 1S-95 standards. A TDMA system may implement a radio technology such as Global System for Mobile Communications (GSM). These various radio technologies and standards are known in the art. W-CDMA and GSM are described in documents from an organization named "3rd Generation Partnership Project" (3GPP). cdma2000 is described in documents from an organization named "3rd Generation Partnership Project 2" (3GPP2). 3GPP and 3GPP2 documents are publicly available. An OFDMA system transmits modulation symbols in the frequency domain on orthogonal frequency subcarriers using orthogonal frequency division multiplexing (OFDM). An SC-FDMA system transmits modulation symbols in the time domain on orthogonal frequency subcarriers. For clarity, these techniques are described below for a M3MO transmission sent in a CDMA system, which may implement W-CDMA and/or cdma2000. [0022] FIG. 1 shows a block diagram of a transmitter 110 and a receiver 150 for a MIMO transmission. For a downlink/forward link transmission, transmitter 110 is part of a base station, and receiver 150 is part of a wireless device. For an uplink/reverse link transmission, transmitter 110 is part of a wireless device, and receiver 150 is part of a base station. A base station is typically a fixed station that communicates with the wireless devices and may also be called a Node B, an access point, and so on. A wireless device may be fixed or mobile and may also be called a user equipment (HE), a mobile station, a user terminal, a subscriber unit, and so on. A wireless device may be a cellular phone, a personal digital assistant (PDA), a wireless modem card, or some other device or apparatus. [0023] At transmitter 110, a transmit (TX) data processor 120 processes (e.g., encodes, interleaves, and symbol maps) traffic data and provides data symbols to multiple (T) CDMA modulators 130a through 130t As used herein, a data symbol is a modulation symbol for data, a pilot symbol is a modulation symbol for pilot, a modulation symbol is a complex value for a point in a signal constellation (e.g., for M-PSK or M-QAM), and pilot is data that is known a priori by both the transmitter and receiver. Each CDMA modulator 130 processes its data symbols and pilot symbols as described below and provides output chips to an associated transmitter unit (TMTR) 136. Each transmitter unit 136 processes (e.g., converts to analog, amplifies, filters, and frequency upconverts) its output chips and generates a modulated signal. T modulated signals from T transmitter units 136a through 136t are transmitted from T antennas 138a through 138t, respectively. [0024] At receiver 150, multiple (R) antennas 152a through 152r receive the trarismitted signals via various signal paths and provide R received signals to R receiver units (RCVR) 154a through 154r, respectively. Each receiver unit 154 processes (e.g., filters, amplifies, frequency downconverts, and digitizes) its received signal and provides samples to a channel and noise estimator 160 and a space-time equalizer 162. Estimator 160 derives channel and noise estimates based on the samples, as described below. Space-time equalizer 162 derives weights based on the channel and noise estimates, performs equalization on the samples with the weights, and provides data chip estimates to T CDMA demodulators (Demod) 170a through 1701 Each CDMA demodulator 170 processes its data chip estimates in a manner complementary to the processing by CDMA modulator 130 and provides data symbol estimates. A receive (RX) data processor 180 processes (e.g., symbol demaps, dcintcrlcavcs, and decodes) the data symbol estimates and provides decoded data. In general, the processing by CDMA demodulator 170 and RX data processor 180 is complementary to the processing by CDMA modulator 130 and TX data processor 120, respectively, at transmitter 110. [0625] Controllers/processors 140 and 190 direct operation of various processing units at transmitter 110 and receiver 150, respectively. MIMOries 142 and 192 store data and program codes for transmitter 110 and receiver 150, respectively. [0026] FIG. 2 shows a block diagram of a CDMA modulator 130 for one transmit antenna. CDMA modulator 130 may be used for each of CDMA modulators 130a through 330t in FIG. 1. CDMA modulator 130 includes a traffic channel processor 210 for each traffic channel used for traffic data and a pilot channel processor 220 for pilot. Within processor 210 for traffic channel m, a spreader 212 spreads data symbols with an orthogonal code cm(k) for traffic channel m. A multiplier 214 scales the output of spreader 212 with a gain gUm and provides data chips x,m (k) for traffic channel m on transmit antenna t. Within pilot channel processor 220, a spreader 222 spreads pilot symbols with an orthogonal code clp(k) for pilot. A multiplier 224 scales the output of spreader 222 with a gain gtJ> and provides pilot chips xtp (k) for the pilot channel on transmit antenna t. A summer 230 sums the chips for all traffic and pilot channcLs. A scrambler 232 multiplies the output of summer 230 with a scrambling sequence s(k} for transmitter 110 and provides output chips xt(k) for transmit antenna t. [0027] The orthogonal codes for the traffic and pilot channels may be orthogonal variable spreading factor (OVSF) codes used in W-CDMA, Walsh codes used in cdma2000, and so on. In an embodiment, T different orthogonal codes are used for the pilot for the T transmit antennas to allow receiver 150 to estimate the channel response for each transmit antenna. The remaining orthogonal codes may be used for each of the T transmit antennas. For this embodiment, the pilot orthogonal codes for the T transmit antennas are distinct whereas the traffic orthogonal codes may be reused for the T transmit antennas. The spreading and scrambling may also be performed in other manners than the manner shown in FIG. 2. [0028] The output chips for each transmit antenna t may be expressed as: (Equation Removed) (l) where x, (fc) is a Kxl vector with K. output chips sent from transmit antenna t, xt (k) is the output chip sent in chip period k from transmit antenna t, crx is the gain for the output chips, and " r" denotes a transpose. K may be set as K = E + L —1, where E is the span of space-time equalizer 162 at receiver 150 and L is the delay spread of the MIMO channel between transmitter 110 and receiver 150. Vector x,(£) would then contain E + L-l chips to be operated on by the space-tune equalizer at the receiver. [0029] In an embodiment, the receiver digitizes the received signal from each receive antenna at V times the chip rate for an oversampling factor of V and obtains V samples for each chip period. The samples for each receive antenna r may be expressed as: (Equation Removed) Eq(2) where yr(k) is a V - E x l vector of samples for receive antenna r, and yr,v (k) is the v_th sample in chip period k for receive antenna r. [0030] The noise observed by the samples in y (k) may be expressed as: (Equation Removed) Eq(3) where nr(_k) is a V- E x 1 noise vector for receive antenna r, and ne,y(k) is the noise observed by yr „(k) . [0031] nr(k) includes channel noise, receiver noise, and interference from other transmitters, which are collectively referred to as "background" noise. As used herein, the term "noise" may refer to just background noise or background noise and inter-strcam interference from other transmit antennas. For example, channel and noise estimation may refer to channel estimation and either background noise estimation or background noise and interference estimation. [0032] The samples for all R receive antennas may be expressed as: (Equation Removed)Eq(4) where Hr, is a V-ExK muitipath chaimel matrix between transmit antenna t and Teceive antenna r. Hr,, may be given as: (Equation Removed) where hr,i,v(l) is a complex channel gain between transmit antenna r and receive antenna r for the l-th channel tap at the v-th sampling time instant. [0033] The impulse response between transmit antenna r and receive antenna r has a length of L chips and may be oversampled at V times chip rate to obtain V • L channel gains. These V-L channel gains may be arranged into V rows such that each row contains L channel gains for one sampling time instant v. Sample yr ,v(k) for sampling time instant v is obtained by convolving the transmitted data chips in x/Ar) with a row of channel gains for that sampling time instant. Hr, contains V - E rows for the V - E samples in y (k). Each row of Hr,, represents the channel impulse response for one sample in y (k). [0034] Equation (4) may be rewritten as follows: (Equation Removed)Eq(6) where y{k)=[y_T(k) y2T(k) ... YRT(k)]T is an R-V-Exl overall received sample vector for all R receive antennas, x(k) = [X1T (k) X2T(k) ... XTT (k)]T is a T - K x 1 overall transmitted chip vector for all T transmit antennas, n(k) = [nT1(k) nT2(k) ... DTR(k)]T is an R-V-Exl overall noise vector for all R receive antennas, and H is an R-V-ExT-K overall channel response matrix. [0035] The overall channel response matrix H may be given as: (Equation Removed)Eq(7) The first K columns of H are for transmit antenna 1, the next K colximns of H are for transmit antenna 2, and so on, and the last K columns of H are for transmit antenna T. Each column of H is denoted as h,- and contains the channel gains observed by one transmitted chip x, (k) . [0036] The transmitted chips may be recovered by space-time equalizer 162, as follows: (Equation Removed) Eq(8) where w, is an R - V - E x 1 equalizer weight vector for transmit antenna t, x,(k + D) is an estimate of data chip x,(k 4- D) sent from transmit antenna t, and " B " denotes a conjugate transpose. [0037] For each chip period k, vector y(&) contains samples for all R receive antennas for the E most recent chip periods. These samples contain components of K = E + L -1 most recent data chips sent from each transmit antenna. Vector y(k) is multiplied with wf to obtain an estimate of one data chip from one transmit antenna. T weight vectors Wj through wT may be formed for the T transmit antennas and used to obtain estimates of T data chips sent from the T transmit antennas in one chip period. [0038] In equation (8), D denotes the delay of the space-time equalizer. For each chip period k, the space-time equalizer provides an estimate of data chip xt (k + D) sent D chip periods later, instead of an estimate of data chip xr(k) sent in the k-th chip period. D is determined "by the design of the equalizer and, in general, 1≤D≤(E + L-1). [0039] The weight vectors w, for the space-time equalizer may be derived based on various receiver processing techniques such as a ininimum mean square error (MMSE) technique, a zero-forcing (ZF) technique, a maximal ratio combining (MRC) technique, and so on. Weight matrices may be derived based on the MMSE, ZF, and MRC techniques, as follows: (Equation Removed)Eq (9)Eq (10)Eq(ll) where R„ is a noise covariance matrix and are T . K x R • V - E weight matrices for the MMSE, ZF and MRC techniques, respectively. The weight vectors wf may be obtained from the rows of W_r, WfJ- and Wf„ [0040] The noise covariance matrix R„ may be expressed as: (Equation Removed) Eq (12) where E{ } denotes an expectation operation. R„ is indicative of the background noise statistics and has a dimension of R-V-ExR- V-E. [0041] The channel and noise estimation techniques described herein may be used with various receiver processing techniques. For clarity, the following description is for the MMSE technique. [0042] For the MMSE techndque, the weight vectors w,, for t = l,...,T, may be expressed as: (Equation Removed) where ht,D = h(t-1)K+D is a channel response vector for transmit antenna t at delay D, Ryy is a matrix indicative of the signal, noise and interference statistics, and I is a set that includes 1 through T • K but excludes (t -1) • K + D . [0043] As shown in equation (13), Ryy includes H-HH and R„. H-HH includes the desired signal as well as interference from other transmit antennas. R„ includes the background noise. Hence, Ryy,, includes the desired signal, the background noise, and the interference from other transmit antennas. For the last equality in equation (13), each summation is over the outer products of all columns of H except for the column for the desired signal, which is htD . hlyD mcludes on-time components arriving at the receiver at delay D. These on-time components include (1) on-time signal component from the desired transmit antenna t and (2) on-time inter-stream interference from the other T —1 transmit antennas also arriving at delay D. Equation (13) minimizes E { wH-y(k)-xt(k + D)\2 }, which is the mean square error between the transmitted data chip x,(k + D) and its estimate wf -y(k). [0044] Enhanced weight vectors w, that take into account the despreading effect of the subsequent CDMA demodulator may be expressed as: (Equation Removed)Eq(14) where G is a gain for a traffic channel being received from transmit antenna t, ■ It is a set that includes 1 through T but excludes t, and Q is a set that includes 1 through T • K but excludes (i -1) • K + D, for i = 1,..., T. [0045] For simplicity, equation (14) assumes the use of the same gain G for all transmit antennas and all traffic channels/orthogonal codes/streams. If J orthogonal codes are used for each transmit antenna and if each orthogonal code has a length of SF, then gain G may be given as G = SF/J . In general, different gains may be used for different orthogonal codes and/or different transmit antennas. [0046] As shown in equations (13) and (14), the MMSE weight vectors may be derived based on the channel response vectors h, through hT.K and the noise covariance matrix £» • The channel response vectors may be derived based on a pilot, which may be transmitted using code division multiplexing (CDIvf), time division multiplexing (TDM), frequency division multiplexing (FDM), and so on. For clarity, the following description assumes the use of a CDM pilot transmitted as shown in FIG. 2. The channel response vectors and the noise covariance matrix may be estimated directly or indirectly with various schemes, as described below. [0047] In a first channel and noise estimation scheme, the channel and noise estimates are derived directly from the received samples. For this scheme, the channel response vectors may be estimated as follows: (Equation Removed)Eq(lS) where p, (k) is a pilot chip sent in chip period k from transmit antenna /, Np is the length of the orthogonal code for the pilot, Eq, is the energy per pilot chip, LPFchuuiei denotes an, averaging operation for the channel estimate, and htp is an estimate of σx .ht,D. [0048] The pilot chips for transmit antenna t may be given as: (Equation Removed) Eq(16) where d(j) is a pilot symbol sent in pilot symbol period./, ctp (k) is the orthogonal code for the pilot for transmit antenna tz and s(k) is the scrambling code for the transmitter. [0049] In equation (15), the overall sample vector y(k) is multiplied with the complex conjugated pilot chip pt(k + D) for delay D and accumulated over the length of the pilot orthogonal code to obtain an initial estimate of ht,D. The initial estimate may be filtered over multiple pilot symbols to obtain ht,D, which is an improved estimate of ht,D. The filtering may be performed based on a finite impulse response (FIR) filter, an infinite impulse response (IIR) filter, and so on. For example, a single-tap ICR. filter with a time constant of 0.33, 0.2, or some other value per slot may be used for the channel filter, LPF Chsnnel. For W-CDMA, each slot spans 2560 chips and covers 10 pilot symbols sent with a 256-chip OVSF code. For cdma2000, each slot spans 768 chips and may cover 12 pilot symbols sent with a 64-chip Walsh code. [0050] The signal, noise and interference statistics may be estimated as follows: (Equation Removed) Eq (17) where LPFdoise denotes an averaging operation for Ryy, and Ryy, is an estimate of Jg.Ryy . The noise filter, LPFnoise, may be the same or different from the channel filter, LPFchannel. [0051] The weight vectors may then be derived as follows: (Equation Removed)Eq(18) Equation (18) is equivalent to equation (13) if ht,D and Ryy are accurate estimates of σx.hD and Ryy, respectively. The SNR of the data chip estimates from the space-time equalizer may be expressed as: (Equation Removed) Eq (19) [0052] The first scheme derives a single estimate for Ryy,. This estimate, Ryy , includes components for both summations in the first equality for equation (14). The two components are not accessible to apply the gain G for the desired traffic channel, and the enhanced weight vectors w, are not derived from Ryy. [0053] FIG. 3 shows a block diagram of a channel and noise estimator 160a, which implements the first channel and noise estimation scheme and is an embodiment of channel and noise estimator 160 in FIG. 1. Estimator 160a includes a serial-to-parallel (S/P) converter 310, a channel estimator 320, and a signal, noise and interference statistics estimator 330. S/P converter 310 receives the samples from R receiver units 154 a through 154r and forms vector y(k). [0054] Within channel estimator 320, a pilot correlator 322 correlates y(&) with a pilot sequence p,(k) for each transmit antenna ./ and provides initial channel estimates for all transmit antennas. A channel filter 324 filters the initial channel estimates and provides final channel estimates, ht,d. Pilot correlator 322 and filter 324 perform processing for equation (15). Within estimator 330, a unit 332 computes the outer product of y(A). A noise filter 334 then filters the output of unit 332 and provides the estimated signal, noise and interference statistics, Ryy • Unit 332 and filter 334 perform processing for equation (17). The scaling factor in equation (15) may be accounted for in various units, e.g., in filter 324. [0055] FIG. 4 shows an embodiment of channel estimator 320 in FIG. 3. For this embodiment, pilot correlator 322 in FIG. 3 is implemented with R pilot correlators 322a through 322r and channel filter 324 is implemented with R channel filters 324a through 324r. One set of pilot correlator 322r and channel filter 324r is provided for each receive antenna. [0056] The samples yhv(Jc) through yR,V(A) from the R receive antennas are provided to R pilot correlators 322a through 322r, respectively. Within each pilot correlator 322r, the samples vr,v(fc) are provided to V-E delay elements 410 that are coupled in scries. Each delay clement 410 provides a delay of one sample period. The V • E delay elements 410 provide their delayed samples to V • E multipliers 412. Each multiplier 412 multiplies its delayed sample with the complex conjugated pilot chip pi(k + D) for delay D. V-E accumulators 414 couple to the V-E multipliers 412. , Each accumulator 414 accumulates the output of an associated multiplier 412 over the length of the pilot orthogonal code, or Np chips, and provides an initial channel gain estimate h't,r,v(l) for each pilot symbol period. [0057] Within each channel filter 324r, V • E channel filters (LPFs) 420 couple to the V • E accumulators 414 in an associated pilot correlator 322r. Each filter 420 filters the initial channel gain estimates from an associated accumulator 414 and provides a final channel gain estimate h,t,r,v(l) for each update interval, e.g., each pilot symbol period. An S/P converter 430 receives the final channel gain estimates from filters 420 for all R receive antennas and provides a channel response vector ht D for each update interval. [0058] In general, the processing units shown in FIGS. 3 and 4 may be implemented in various manners. For example, these units may be implemented with dedicated hardware, a shared digital signal processor (DSP), and so on. [005.9] FIG. 5 shows an embodiment of a process 500 for performing channel and noise estimation based on the first scheme. Samples are obtained from multiple receive antennas for a MIMO transmission sent from multiple transmit antennas (block 512). Channel estimates are derived, e.g., by correlating the samples with at least one pilot sequence (block 514). The channel estimates may comprise T channel response vectors h,t,d for t — 1,..., T , for T transmit antennas at delay D. The channel response vector for each transmit antenna may be obtained based on a pilot sequence for that transmit antenna. The signal, noise and interference statistics are estimated based on the samples, e.g., by computing cross product of the samples and filtering the cross product results (block 516). The channel estimates may be derived with a first filter having a first bandwidth. The estimated signal, noise, and interference statistics may be derived with a second filter having a second bandwidth. The first and second filter bandwidths may be the same or different and may be fixed or configurable, e.g., adjusted based on channel conditions. Equalizer weights arc derived based on the channel estimates and the estimated signal, noise and interference statistics (block 518). The samples are filtered with the equalizer weights to obtain estimates of the data chips sent from the transmit antennas (block 520). [0060] In a second channel and noise estimation scheme, the noise estimation is performed based on an estimate of the total received energy, Tσ, at the receiver and an estimate of the signal and interference energy, lor- The terms "energy" and "power" are often used interchangeably. [0061] For the second scheme, the elements of vectors σx h1 through σx.hT.K may be estimated as follows: (Equation Removed)Eq (20) where Ar,t,v(l) is an estimate of channel gain hr,t,v{l). Channel gain estimates may be obtained for r = l, ..., R, r = l, ..., T, v = l, ..., V, and / = 0, ..., L-l. Equation (20) is equivalent to equation (15). However, hr,t,v(l) may be computed for different values of r, t, v and / with equation (20) whereas h,D may be computed for different values of t and a specific value of D with equation (15). The channel gain estimates hr,t,v{l) may be used to form h,, which is an estimate of h,, for i = 1, ..., T • K [0062] For the second scheme, the background noise may be assumed to be spatio-temporally white, so that Rn = σn2.I,, where σ2n is the noise variance and I is an identity matrix. The noise variance may be estimated as follows: (Equation Removed) Eq(21) where I is the estimated total received energy, and Ior is the estimated signal and interference energy. [0063J In an embodiment, It and Iar may be derived as follows: (Equation Removed) [0064] In another embodiment, Jor may be derived by stmiming channel gains with sufficient energy. A channel gain may be considered to be sufficiently strong if, e.g., hr,t,v(l)2≥Eth, where Eth is a threshold. Eth may be a fixed value or a configurable value that may be derived based on the total energy for all channel gains. [0065] The noise covariance matrix may then be estimated as follows: (Equation Removed)Eq(24) where R„ is an estimate of Rn. Rn and h, may be used to derive w, as shown in equation (13) or w, as shown in equation (14). [0066] FIG. 6 shows a block diagram of a channel and noise estimator 160b, which implements the second channel and noise estimation scheme and is another embodiment of channel and noise estimator 160 in FIG. 1. Estimator 160b includes a multiplexer (Mux) 610, a channel estimator 620, and a noise estimator 630. Multiplexer 610 receives the samples from R receiver units 154a through 154r and provides a stream of samples yr,v(k) in a desired order. [0067] Within channel estimator 620, a pilot correlator 622 correlates yr v(k) with a pilot sequence pt(k) for each transmit antenna and provides initial channel gain estimates for all transmit antennas. A channel filter 624 filters the initial channel gain estimates and provides final channel gain estimates, Ar,t,v(/). Pilot correlator 622 and filter 624 perform the processing shown in equation (20). An S/P converter 626 receives the final channel gain estimates for all transmit antennas and provides channel response vectors h,-, for / = 1, ..., T • K. [0068] Noise estimator 630 estimates the total received energy I0, the signal and interference energy Ior, and the noise For the I0 estimate, an energy computation unit 642 computes the energy of each sample. An accumulator 644 accumulates the sample energies across the V sample periods and the R receive antennas and provides an initial I0 estimate for each chip period. A noise filter 646 filters the initial I0 estimates and provides the final I0 estimate, I0. For the Ior estimate, an energy computation unit 652 computes the energy of each channel tap as \| kt (/) \|2. An accumulator 654 accumulates the channel tap energies across the V sample periods, the L channel taps, the T transmit antennas, and the R receive antennas and provides an initial Ior estimate. The accumulation may also be performed over channel taps with sufficient energy. A noise filter 656 filters the initial /„. estimates and provides the final Ior estimate, Ior. For the σ2 estimate, a summer 658 subtracts Iar from /„ and provides the noise variance estimate, Σ2 . A unit 660 forms a noise covariance matrix R, based on ΣN„2 , as shown in equation (24). The scaling factors in equations (20), (22) and (23) maybe accounted for in various units, e.g., in filters 624, 646 and 656. [0069] FIG. 7 shows an embodiment of a process 700 for performing channel and noise estimation based on the second scheme. Samples are obtained from multiple receive antennas for a MIMO transmission sent from multiple transmit antennas (block 712). Channel estimates are derived, e.g., by correlating the samples with at least one pilot sequence (block 714). The total received energy I0 is estimated based, on the samples, e.g., by computing the energy of each sample, accumulating the sample energies across the receive antennas and for a given time interval, and filtering the accumulated results for different time intervals (block 716). The signal and interference energy Ior is estimated based on the samples, e.g., by computing the energy of each channel tap in the channel estimate, accumulating the energies of all or sufficiently strong channel taps for a given time interval, and filtering the accumulated results for different time intervals (block 718). The channel estimates may be derived with a first filter having a first bandwidth. The total received energy as well as the signal and interference energy may be estimated with a second filter having a second bandwidth that may be the same or different from the first bandwidth. The noise σn2 Ls estimated based on the estimated total received energy and the estimated signal and interference energy (block 720). Equalizer weights are derived based on the channel estimates and the estimated noise (block 722). The samples are filtered with the equalizer weights to obtain estimates of the data chips sent from the transmit antennas (block 724). [0070] The second scheme can provide good performance for operating scenarios in which the background noise is spatio-tcmporally white. The second scheme also works well for a severe multipath environment since (a) the background noise Rn is less significant and (b) the dominant component of Ryy is likely to be multipath interference that may be accurately estimated based on the channel estimates obtained from equation (20). [0071] In a third channel and noise estimation scheme, Ryy is decomposed into on-time components and remaining components, which are estimated separately. For the third scheme, initial channel estimates may be derived as follows: (Equation Removed) Eq (25) where hr,d (j) is an estimate of hr,d for pilot symbol period j. [0072] The channel response vectors may then be estimated as follows: (Equation Removed)Eq (26) Equations (25) and (26) are equivalent to equation (15). [0073] Equations (13) and (14) for the equalizer weights may be rewritten as follows: (Equation Removed) where Rat is a matrix indicative of signal and on-time interference statistics, and R,. is a matrix indicative of noise and multipath interference statistics. [0074] Rot may be expressed as: (Equation Removed) Rot includes the desired signal as well as the on-time interference from the other transmit antennas. [0075] Rc may be expressed as: (Equation Removed) where g is a set that includes 1 through TK but excludes (/-1)K + D, for i = 1,..., T. Rc includes (1) multipath interference, which is interference that is not on-time and represented by the summation in equation (30), and (2) background noise, which is represented by Rn.. Rc,. does not include contribution from the signal and on-time interference, which is represented by ht,D. Hence, Rc may be estimated by performing differential-based multipath plus noise estimation. [0076] To estimate Rc, ht,D(j) is first differentiated as follows: (Equation Removed) Eq(31) The differential operation in equation (31) cancels the on-time components in ht,D(j) if the channel response remains constant over two consecutive pilot symbol periods. A scaling factor of 1 / is introduced in equation (31) so that the differential operation does not change the second-order moment of the multipath interference and the background noise. An estimate of Rc may then be derived as follows: (Equation Removed) Rc docs not include contributions from the on-time components, which arc removed by the differential operation, in equation (31). [0077] An estimate of Rot may be derived as follows: (Equation Removed) Eq(33) [0078] For the weight vectors shown in equation (13) and (27), Rc may be augmented with Rot to obtain an estimate of Ryy,, as follows: (Equation Removed) Eq(34) The weight vectors may then be derived as: (Equation Removed) Eq(35) [0079] For the enhanced weight vectors shown in equation (14) and (28), Rc may be augmented with Rot, to obtain an estimate of Ryy, as follows: (Equation Removed) Eq (36) The enhanced weight vectors may then be derived as: (Equation Removed) Eq(37) [0080] FIG. 8 shows a block diagram of a channel and noise estimator 160c, which implements the third channel and noise estimation scheme and is yet another embodiment of channel and noise estimator 160 in FIG. 1. Estimator 160c includes an S/P converter 810, a channel estimator 820, and a signal, noise and interference statistics estimator 830. S/P converter 810 receives the samples from R receiver units 154a through 154r and forms vector y(k). [0081] Within channel estimator 820, a pilot correlator 822 correlates y(k) with a pilot sequence ps(k) for cach transmit antenna t and provides an initial channel estimate, ht,D(j) for each transmit antenna in each pilot symbol period J. A channel filter 824 filters the initial channel estimates and provides final channel estimates, hc,D . Pilot correlator 822 performs the processing shown in equation (25). Filter 824 performs the processing shown in equation (26). [0082] Within estimator 830, a unit 832 receives and differentiates the initial channel estimates, ht,D (J). as shown in equation (31) and provides the difference, Δht,D(j). A unit 834 computes the outer product of Δht,D(j) • -A- unit 836 sums the output of unit 834 across all T transmit antennas. A noise filter 838 filters the output from unit 836 and provides the estimated noise and raultipath interference statistics, Rc. Units 834, 836 and 838 perform the processing shown in equation (32). A unit 844 computes the outer product of ht,D . A unit 846 sums the output of unit 844 across all T transmit antennas and provides the estimated signal and on-time interference statistics, Rot . Units 844 and 846 perform the processing shown in equation (33). A matrix summer 850 sums Rc. with Rot, and provides the estimated signal, noise and interference statistics, Ryy. The scaling factors in equations (26), (31) and (32) may be accounted for in various units, e.g., in filters 824 and 838. [0083] Rc. and Rot may also be estimated in other manners. For example, an estimate of Rot, may be obtained by computing the outer product of ht,D(J), surorning across T transmit antennas, and filtering the summed results. As another example, an estimate of Rc may be obtained by filtering Δht,d(J), computing the outer product of the filtered Δht,d(/) , and summing across the T transmit antennas. [0084] FIG. 9 shows an embodiment of a process 900 for performing channel and noise estimation based on the third scheme. Samples are obtained from multiple receive antennas for a MIMO transmission sent from multiple transmit antennas (block 912). First or initial channel estimates may be derived, e.g., by correlating the samples with at least one pilot sequence (block 914). The first channel estimates are filtered with a first filter to obtain second or final channel estimates (block 916). [0085] The signal and on-time interference statistics are estimated based on the samples, e.g., by computing the cross-product of the second channel estimates and summing across the transmit antennas (block 918). The noise and multipath interference statistics are also estimated based on the samples, e.g., by differentiating the first channel estimates, computing cross-product of the differentiated results, and filtering the cross-product results with a second filter (block 920). The first and second filters may have the same or different bandwidths, which may be fixed or configurable, e.g., adjusted based on channel conditions. The signal, noise and interference statistics are then estimated based on the estimated signal and on-time interference statistics and the estimated noise and multipath mtcrfcrcncc statistics (block 922). [0086] Equalizer weights are derived based on the second channel estimates and the estimated signal, noise and interference statistics (block 924). The samples are filtered with the equalizer weights to obtain estimates of the data chips sent from the transmit antennas (block 926). [0087] As shown in equation (26), the on-time components at delay D may be averaged by the channel LPF channel, to obtain ht,D, which is used to derive Rol. The background noise and multipath interference may be averaged by the noise filter, LPFnoisc, to obtain Rc. The use of two filters to average the desired signal in hjD and the noise and multipath interference in Rc may be beneficial in some environments. For example, in a single-path high-velocity channel, a small bandwidth or large time constant may be used for the noise filter to improve the accuracy of Rc, and a large bandwidth or small time constant may be used for the channel filter to prevent utilization of stale or obsolete channel estimates h, D. The noise and multipath interference estimation is typically more accurate with a smaller bandwidth for the noise filter. A smaller noise filter bandwidth is beneficial for a severe multipath channel in order to sufficiently suppress large uncorrelated components associated with multipath. However, the bandwidth of the noise filter should not be reduced too much in a severe multipath high-velocity channel since multipath component becomes the dominant component that determines Rc, and the multipath component may become outdated if the noise filter bandwidth is too small. [0088] The third scheme provides good performance for many channel environments, especially for a single-path high geometry channel. The use of the robust diffcrcntial-bascd noise estimation allows,the third scheme to work well in various channel conditions. The third scheme also has lower computational complexity than the first scheme since the outer products in equations (32) and (33) are performed every pilot symbol period instead of every chip period. The third scheme also provides good performance even with a colored background noise. [0089] For the third scheme, the differential operation in equation (31) assumes that the channel is constant over two consecutive pilot symbol periods. A sufficiently short pilot symbol duration may be used for a high-velocity channel so that the noise and multipath interference estimation is not outdated. Computer simulation results show that a pilot symbol duration of 256 chips or less for HSDPA in W-CDMA provides good performance with 30km/h velocity. [0090] In general, the estimation of the multipath interference in Rc may be improved with more vectors for h't,DO)- The number of h't,D(j) vectors may be increased by taking partial integration of the pilot symbols. For example, if the pilot symbol duration is 512 chips long, then the integration length may be set to 128 chips, and four ht,D (j) vectors may be obtained per pilot symbol. The partial integration may result in crosstalk between orthogonal pilots from different transmit antennas. However, since the orthogonal pilot patterns and the channel estimates h,t,D are known, the pilot crosstalk may be estimated and subtracted out. This approach may still suffer from the leakage (crosstalk) of some OCNS channels if they have a symbol length greater than 128, but it can be acceptable if the power of the corresponding OCNS is small. Alternatively, in order to maximize the performance, we may take the length of pilot symbols shorter than that of some OCNS symbols in the system design. [0091] In a fourth channel and noise estimation scheme, multiple estimation schemes are supported, and one estimation scheme is selected for use based on channel conditions. In an embodiment, the second scheme is selected for a severe multi-path high-velocity channel and the third scheme is selected for a single-path high-geometry channel. Other combinations of channel and noise estimation schemes may also be used. [0092] The channel conditions may be detected in various manners and with various metrics. In an embodiment, the channel conditions are characterized by delay spread. For this embodiment, a searcher may search for signal paths or multipaths, e.g., similar to the search performed for a CDMA receiver. The delay spread may be computed as the difference between the earliest and latest arriving signal paths at the receiver. In another embodiment, the channel conditions are characterized by a path energy ratio. For this embodiment, the energy of the strongest signal path is computed, and the combined energy of all other signal paths identified as significant (e.g., with energy exceeding a threshold Eth) is also computed. The path energy ratio is the ratio of the combined energy for the significant paths to the energy for the strongest path. A low padi energy ratio may indicate a single path environment whereas a high path energy ratio may indicate a multipath environment. In yet another embodiment, the velocity of the receiver is estimated, e.g., based on an estimate of Doppler and/or some other velocity-indicating parameter. [0093] In an embodiment, the third estimation scheme is selected if the delay spread is smaller than a delay threshold Dth or if the path energy ratio is less than a predetermined threshold Pa, or if the velocity is below a velocity threshold Vth. The second estimation scheme may be selected otherwise. If the velocity is low, then the bandwidth of the noise filter may be reduced to improve noise estimation. In general, the third estimation scheme may be used if a sufficient number of noise and multipath interference samples that are relatively current can be obtained, e.g.; via a short pilot symbol period or small bandwidth for the noise filter. Otherwise, the second estimation scheme may be used. [0094] FIG. 10 shows an embodiment of a process 1000 for performing channel and noise estimation based on the fourth scheme. Samples are obtained from multiple receive antennas for a MIMO transmission sent from multiple transmit antennas (block 1012). Channel conditions arc detennined based on the samples (block 1014). The channel conditions may be characterized by (1) delay spread, which may be estimated based on the earliest and latest arriving signal paths, (2) single path or multipath environment, which may be determined based on the energies of the signal paths detected at the receiver, (3) velocity, which may be estimated based on Doppler, and/or (4) possibly other criteria. 10095] A channel and noise estimation scheme is selected from among multiple channel and noise estimation schemes based on the channel conditions (block 1016). The multiple channel and noise estimation schemes may include all or any combination of the schemes described above. For example, the second and third channel and noise estimation schemes described above may be supported. The third channel and noise estimation scheme may be selected if the channel conditions indicate a small delay spread, a single path environment, low velocity, or a combination thereof. The second channel and noise estimation scheme may be selected if the channel conditions indicate a large delay spread, a multipath environment, high velocity, or a combination thereof. Channel and noise estimation is then performed based on the selected channel and noise estimation scheme (block 1018). [0096] The channel and noise estimation techniques described herein may be implemented by various means. For example, these techniques may be implemented in hardware, firmware, software, or a combination thereof. For a hardware implementation, the processing units used to perform channel and noise estimation may be implemented within one or more application specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field programmable gate arrays (FPGAs), processors, controllers, micro-controllers, microprocessors, electronic devices, other electronic units designed to perform the functions described herein, or a combination thereof. [0097] For a firmware and/or software implementation, the techniques may be implemented with modules (e.g., procedures, functions, and so on) that perform the functions described herein. The firmware and/or software codes may be stored in a MIMOry (e.g., MIMOry 192 in FIG. 1) and executed by a processor (e.g., processor 190). The MIMOry may be implemented within the processor or external to the processor, in which case it can be communicatively coupled to the processor via various means as is known in the art. [0098] 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. CLAIMS 1. An apparatus comprising: at least one processor to obtain samples from multiple receive antennas for a multiple-input multiple-output (MEMO) transmission sent from multiple transmit antennas, to derive channel estimates by correlating the samples with at least one pilot sequence, and to estimate signal, noise and interference statistics based on the samples; and a memory coupled to the at least one processor. 2. The apparatus of claim 1, wherein the at least one processor derives the channel estimates with a first filter having a first bandwidth, and estimates the signal, noise and interference statistics with a second filter having a second bandwidth. 3. The apparatus of claim 1, wherein the at least one processor derives equalizer "weights based on the channel estimates and the estimated signal, noise and mterference statistics, and filters the samples with the equalizer weights. 4. The apparatus of claim 1, wherein multiple signals are transmitted from the multiple transmit antennas, and wherein each signal comprises multiple data streams multiplexed with different orthogonal codes. 5. A method comprising: obtaining samples from multiple receive antennas for a multiple-input multiple-output (MIMO) transmission sent from multiple transmit antennas; deriving channel estimates by correlating the samples with at least one pilot sequence; and estimating signal, noise and interference statistics based on the samples. 6. The method of claim 5, wherein the deriving the channel estimates comprises deriving the channel estimates with a first filter having a first bandwidth, and wherein the estimating the signal, noise and interference statistics comprises estimating the signal, noise and interference statistics with a second filter having a second bandwidth. 7. The method of claim 5, further comprising: deriving equalizer weights based on the channel estimates and the estimated signal, noise and interference statistics; and filtering the samples with the equalizer weights. 8. An apparatus comprising: means for obtaining samples from multiple receive antennas for a multiple-input multiple-output (MIMO) transmission sent from multiple transmit antennas; means for deriving channel estimates by correlating the samples with at least one pilot sequence; and means for estimating signal, noise and interference statistics based on the samples. 9. The apparatus of claim 8, wherein the means for deriving the channel estimates comprises means for deriving the channel estimates with a first filter having a first bandwidth, and wherein the means for estimating the signal, noise and interference statistics comprises means for estimating the signal, noise and interference statistics with a second filter having a second bandwidth. 10. The apparatus of claim 8, further comprising: means for deriving equalizer weights based on the channel estimates and the estimated signal, noise and interference statistics; and means for filtering the samples with the equalizer weights. 11. An apparatus comprising: at least one processor to obtain samples from multiple receive antennas for a multiple-input multiple-output (MIMO) transmission sent from multiple transmit antennas, to estimate total received energy based on the samples, to estimate signal and interference energy based on the samples, and to estimate noise based on the estimated total received energy and the estimated signal and interference energy; and a memory coupled to the at least one processor. 12. The apparatus of claim 11, wherein the at least one processor derives channel estimates by correlating the samples with at least one pilot sequence, and estimates the signal and interference energy based on the channel estimates. 13. The apparatus of claim 12, wherein the at least one processor identifies channel taps with sufficient strength from among all channel taps for the channel estimates, and determines the signal and interference energy based on the identified channel taps. 14. The apparatus of claim 11, wherein the at least one processor computes energy of each of the samples, accumulates energies of the samples across the multiple receive antennas, and filters accumulated results for different time intervals to obtain the estimated total received energy. 15. The apparatus of claim 12, wherein the at least one processor computes energy of each of multiple channel taps for the channel estimates, accumulates energies of the multiple channel taps, and filters accumulated results for different time intervals to obtain the estimated signal and interference energy. 16. The apparatus of claim 12, wherein the at least one processor derives the channel estimates with a first filter having a first bandwidth, and estimates the total received energy and the signal and interference energy with a second filter having a second bandwidth. 17. The apparatus of claim 16, wherein the at least one processor selects the first and second bandwidths based on channel conditions. 18. A method comprising: obtaining samples from multiple receive antennas for a multiple-input multiple-output (MEMO) transmission sent from multiple transmit antennas; estimating total received energy based on the samples; estimating signal and interference energy based on the samples; and estimating noise based on the estimated total received energy and the estimated signal and interference energy. 19. The method of claim 18, wherein the estimating the total received energy comprises computing energy of each of the samples, and accumulating energies of the samples. 20. The method of claim 18, wherein the estimating the signal and interference energy comprises deriving channel estimates comprising multiple channel taps, computing energy of each of the multiple channel taps, and accumulating energies of the multiple channel taps. 21. An apparatus comprising: means for obtaining samples from multiple receive antennas for a multiple-input multiple-output (MTMO) transmission sent from multiple transmit antennas; means for estimating total received energy based on the samples; means for estimating signal and interference energy based on the samples; and means for estimating noise based on the estimated total received energy and the estimated signal and interference energy. 22. The apparatus of claim 21, wherein the means for estimating the total received energy comprises means for computing energy of each of the samples, and means for accumulating energies of the samples. 23. The apparatus of claim 21, wherein the means for estimating the signal and interference energy comprises means for deriving channel estimates comprising multiple channel taps, means for computing energy of each of the multiple channel taps, and means for accumulating energies of the multiple channel taps. 24. An apparatus comprising: at least one processor to obtain samples from multiple receive antennas for a multiple-input multiple-output (MEMO) transmission sent from multiple transmit antennas, to estimate signal and on-time interference statistics based on the samples, to estimate noise and multipath interference statistics based on the samples, and to estimate signal, noise and interference statistics based on the estimated signal and on-time interference statistics and the estimated noise and multipath interference statistics; and a memory coupled to the at least one processor. 25. The apparatus of claim 24, wherein the at least one processor derives channel estimates by correlating the samples with at least one pilot sequence. 26. The apparatus of claim 24, wherein the at least one processor derives first channel estimates based on the samples, and filters the first channel estimates to obtain second channel estimates. 27. The apparatus of claim 26, wherein the at least one processor differentiates the first channel estimates to obtain differentiated results, and estimates the noise and multipath interference statistics based on the differentiated results. 28. The apparatus of claim 26, wherein the at least one processor estimates the signal and cm-time interference statistics based on the second channel estimates. 29. The apparatus of claim 24, wherein the at least one processor estimates the signal and on-time interference statistics with a first filter having a first bandwidth, and estimates the noise and multipath interference statistics with a second filter having a second bandwidth. 30. The apparatus of claim 29, wherein the at least one processor selects the first and second bandwidths based on channel conditions. 31. A method comprising: obtaining samples from multiple receive antennas for a multiple-hiput multiple-output (MTMO) transmission sent from multiple transmit antennas; estimating signal and on-time interference statistics based on the samples; estimating noise and multipath interference statistics based on the samples; and estimating signal, noise and interference statistics based on the estimated signal and on-time interference statistics and the estimated noise and multipath interference statistics. 32. The method of claim 31, further comprising: deriving first channel estimates based on the samples; and filtering the first channel estimates to obtain second channel estimates. 33. The method of claim 32, wherein the estimating the noise and multipath interference statistics comprises differentiating the first channel estimates to obtain differentiated results, and estimating the noise and multipath interference statistics based on the differentiated results. 34. The method of claim 32, wherein the estimating the signal and on-time interference statistics comprises estimating the signal and on-time interference statistics based on the second channel estimates. 35. An apparatus comprising: means for obtaining samples from multiple receive antennas for a multiple-input multiple-output (MIMO) transmission sent from multiple transmit antennas; means for estimating signal and on-time interference statistics based on the samples; means for estimating noise and multipath interference statistics based on the samples; and means for estimating signal, noise and interference statistics based on the estimated signal and on-time interference statistics and the estimated noise and multipath interference statistics. 36. The apparatus of claim 35, further comprising: means for deriving first channel estimates based on the samples; and means for filtering the first channel estimates to obtain second channel estimates. 37. The apparatus of claim 36, wherein the means for estimating the noise and multipafh interference statistics comprises means for differentiating the first channel estimates to obtain differentiated results, and means for estimating the noise and multipath interference statistics based on the differentiated results. 38. The apparatus of claim'36, wherein the means for estimating the signal and on-time interference statistics comprises means for estimating the signal and on-time interference statistics based on the second channel estimates. 39. An apparatus comprising: at least one processor to obtain samples from multiple receive antennas for a multiple-input multiple-output (MEMO) transmission sent from multiple transmit antennas, to determine channel conditions based on the samples, to select one of multiple channel and noise estimation schemes based on the channel conditions, and to perform channel and noise estimation based on the selected channel and noise estimation scheme; and a memory coupled to the at least one processor. 40. The apparatus of claim 39, wherein for the selected channel and noise estimation scheme the at least one processor derives channel estimates based on the samples, estimates signal and on-time interference statistics based on the samples, estimates noise and multipath interference statistics based on the samples, and estimates signal, noise and interference statistics based on the estimated signal and on-time interference statistics and the estimated noise and multipath mterference statistics. 41. The apparatus of claim 40, wherein the selected channel and noise estimation scheme is used if the channel conditions indicate a small delay spread, a single path environment, low velocity, or a combination thereof. 42. The apparatus of claim 39, wherein for the selected channel and noise estimation scheme the at least one processor derives channel estimates based on the samples, estimates total received energy based on the samples, estimates signal and interference energy based on the channel estimates, and estimates noise based on the estimated total received energy and the estimated signal and interference energy. 43. The apparatus of claim 42, wherein the selected channel and noise estimation scheme is used if the channel conditions indicate a large delay spread, a multipath environment, high velocity, or a combination thereof. 44. The apparatus of claim 39, wherein for the selected channel and noise estimation scheme the at least one processor derives channel estimates based on the samples and estimates signal, noise and interference statistics based on the samples. 45. The apparatus of claim 39, wherein the at least one processor determines delay spread in a communication channel between the multiple transmit antennas and the multiple receive antennas, and selects the channel and noise estimation scheme based on the delay spread. 46. The apparatus of claim 39, wherein the at least one processor determines energies of signal paths in a communication channel between the multiple transmit antennas and the multiple receive antennas, and selects the channel and noise estimation scheme based on the energies of the signal paths. 47. The apparatus of claim 39, wherein the at least one processor selects the channel and noise estimation scheme based on estimated velocity. 48. A method comprising: obtaining samples from multiple receive antennas for a multiple-input multiple-output (MIMO) transmission sent from multiple transmit antennas; determining channel conditions based on the samples; selecting one of multiple channel and noise estimation schemes based on the channel conditions; and performing channel and noise estimation based on the selected channel and noise estimation scheme. 49. The method of claim 48, wherein the performing channel and noise estimation comprises deriving channel estimates based on the samples, estimating signal and on-time interference statistics based on the samples, estimating noise and multipath interference statistics based on the samples, and estimating signal, noise and interference statistics based on the estimated signal and on-time interference statistics and the estimated noise and multipath interference statistics. 50. The method of claim 48, wherein the performing channel and noise estimation comprises deriving channel estimates based on the samples, estimating total received energy based on the samples, estimating signal and interference energy based on the channel estimates, and estimating noise based on the estimated total received energy and the estimated signal and interference energy. 51. The method of claim 48, wherein the selecting one of the multiple channel and noise estimation schemes comprises selecting one of the multiple channel and noise estimation schemes based on delay spread, path conditions, velocity, or a combination thereof. 52. An apparatus comprising: means for obtaining samples from multiple recerve antennas for a multiple-input multiple-output (MEMO) transmission sent from multiple transmit antennas; means for deterniining channel conditions based on the samples; means for selecting one of multiple channel and noise estimation schemes based on the channel conditions; and means for performing channel and noise estimation based on the selected channel and noise estimation scheme. 53. The apparatus of claim 52, wherein the means for performing channel and noise estimation comprises means for deriving channel estimates based on the samples, means for estimating signal and on-time interference statistics based on the samples, means for estimating noise and multipath interference statistics based on the samples, and means for estimating signal, noise and interference statistics based on the estimated signal and on-time interference statistics and the estimated noise and multipath interference statistics. 54. The apparatus of claim 52, wherein the means for performing channel and noise estimation comprises means for deriving channel estimates based on the samples, means for estimating total received energy based on the samples, means for estimating signal and interference energy based on the channel estimates, and means for estimating noise based on the estimated total received energy and the estimated signal and interference energy. 55. The apparatus of claim 52, wherein the means for selecting one of the multiple channel and noise estimation schemes comprises means for selecting one of the multiple channel and noise estimation schemes based on delay spread, path conditions, velocity, or a combination thereof.

Full Text

METHOD AND APPARATUS FOR CHANNEL AND NOISE
ESTIMATION
1. Claim of Priority under 35 U.S.C. §119
[0001] The present Application for Patent claims priority to Provisional Application Serial No. 60/731,423, entitled "METHOD AND APPARATUS FOR SPACE-TIME EQUALIZATION IN WIRELESS COMMUNICATIONS," filed October 28, 2005, assigned to the assignee hereof, and expressly incorporated herein by reference.
BACKGROUND
I. Field
[0002] The present disclosure relates generally to communication, and more specifically to techniques for performing channel and noise estimation for a multiple-input multiple-output (MEMO) transmission.
II. Background
[0003] A MIMO transmission is a transmission from multiple (T) transmit antennas to multiple (R) receive antennas. For example, a transmitter may simultaneously transmit T data streams from the T transmit antennas. These data streams are distorted by the propagation environment and further degraded by noise. A receiver receives the transmitted data streams via the R receive antennas. The received signal from each receive antenna contains scaled versions of the transmitted data streams, possibly at different delays determined by the propagation environment. The transmitted data streams are thus dispersed among the R received signals from the R receive antennas. The receiver would then perform receiver spatial processing (e.g., space-time equalization) on the R received signals in order to recover the transmitted data streams. [0004] The receiver may derive channel and noise estimates for the MIMO channel and may then derive weights for a space-time equalizer based on the channel and noise estimates. The quality of the channel and noise estimates may have a large impact on performance. There is therefore a need in the art for techniques to derive quality channel and noise estimates for a MIMO transmission.
SUMMARY
[0005] According to an embodiment of the invention, an apparatus is described
which includes at least one processor and a memory. The processors) obtain samples from multiple receive antennas for a MIMO transmission sent from multiple transmit antennas, derive channel estimates by correlating the samples with at least one pilot sequence, and estimate signal, noise and interference statistics based on the samples. The MIMO transmission may comprise multiple modulated signals sent from the multiple transmit antennas. Each modulated signal may comprise multiple data streams multiplexed with different orthogonal codes.
[0006] According to another embodiment, an apparatus is described which includes at least one processor and a memory. The processor's) obtain samples from multiple receive antennas for a MIMO transmission sent from multiple transmit antennas, estimate total received energy based on the samples, estimate signal and interference energy based on the samples, and estimate noise based on the estimated total received energy and the estimated signal and interference energy.
[0007] According to yet another embodiment, an apparatus is described which
includes at least one processor and a memory. The processor's) obtain samples from multiple receive antennas for a MIMO transmission sent from multiple transmit antennas, estimate signal and on-time interference statistics based on the samples, estimate noise and multipath interference statistics based on the samples, and estimate signal, noise and interference statistics based on the estimated signal and on-time interference statistics and the estimated noise and multipath interference statistics. The on-time interference is interference that arrives at the same time as a desired signal, and the multipath interference is interference that is not on time.
[0008] According to yet another embodiment, an apparatus is described which
includes at least one processor and a memory. The processor's) obtain samples from
multiple receive antennas for a MIMO transmission sent from multiple transmit
antennas, determine the channel conditions based on the samples, select one of multiple
channel and noise estimation schemes based on the channel conditions, and perform
channel and noise estimation based on the selected channel and noise estimation
scheme. To determine the channel conditions, the processors) may process the samples
to determine delay spread, to identify single path or multipath environment, and so on.
[0009] Various aspects and embodiments of the invention are described in further
detail below.
BRIEF DESCRIPTION OF THE DRAWINGS
[0010] FIG. 1 shows a transmitter and a receiver for a MIMO transmission.
[0011] FIG. 2 shows a CDMA modulator for one transmit antenna.
[0012] FIG. 3 shows a channel and noise estimator for a first scheme.
10013J FIG. 4 shows a channel estimator.
[0014] F7G. 5 shows a process for the first channel and noise estimation scheme.
[0015] FIG. 6 shows a channel and noise estimator for a second scheme.
[0016] FIG. 7 shows a process for the second channel and noise estimation scheme.
[0017] FIG. 8 shows a channel and noise estimator for a third scheme.
[0018] FIG. 9 shows a process for the third channel and noise estimation scheme.
[0019] FIG. 10 shows a process for a fourth channel and noise estimation scheme.
DETAILED DESCRIPTION
[0020] The word "exemplary" is used herein to mean "serving as an example, instance, or illustration." Any embodiment described herein as "exemplary" is not necessarily to be construed as preferred or advantageous over other embodiments. [0021] The channel and noise estimation techniques described herein may be used for various communication systems such as Code Division Multiple Access (CDMA) systems, Time Division Multiple Access (TDMA) systems, Frequency Division Multiple Access (FDMA) systems, Orthogonal FDMA (OFDMA) systems, Single-Carrier FDMA (SC-FDMA) systems, and so on. A CDMA system may implement one or more radio technologies such as Wideband-CDMA (W-CDMA), cdma2000, and so on. cdma2000 covers 1S-2000, IS-856, and 1S-95 standards. A TDMA system may implement a radio technology such as Global System for Mobile Communications (GSM). These various radio technologies and standards are known in the art. W-CDMA and GSM are described in documents from an organization named "3rd Generation Partnership Project" (3GPP). cdma2000 is described in documents from an organization named "3rd Generation Partnership Project 2" (3GPP2). 3GPP and 3GPP2 documents are publicly available. An OFDMA system transmits modulation symbols in the frequency domain on orthogonal frequency subcarriers using orthogonal frequency division multiplexing (OFDM). An SC-FDMA system transmits modulation symbols in the time domain on orthogonal frequency subcarriers. For clarity, these techniques are
described below for a M3MO transmission sent in a CDMA system, which may implement W-CDMA and/or cdma2000.
[0022] FIG. 1 shows a block diagram of a transmitter 110 and a receiver 150 for a MIMO transmission. For a downlink/forward link transmission, transmitter 110 is part of a base station, and receiver 150 is part of a wireless device. For an uplink/reverse link transmission, transmitter 110 is part of a wireless device, and receiver 150 is part of a base station. A base station is typically a fixed station that communicates with the wireless devices and may also be called a Node B, an access point, and so on. A wireless device may be fixed or mobile and may also be called a user equipment (HE), a mobile station, a user terminal, a subscriber unit, and so on. A wireless device may be a cellular phone, a personal digital assistant (PDA), a wireless modem card, or some other device or apparatus.
[0023] At transmitter 110, a transmit (TX) data processor 120 processes (e.g., encodes, interleaves, and symbol maps) traffic data and provides data symbols to multiple (T) CDMA modulators 130a through 130t As used herein, a data symbol is a modulation symbol for data, a pilot symbol is a modulation symbol for pilot, a modulation symbol is a complex value for a point in a signal constellation (e.g., for M-PSK or M-QAM), and pilot is data that is known a priori by both the transmitter and receiver. Each CDMA modulator 130 processes its data symbols and pilot symbols as described below and provides output chips to an associated transmitter unit (TMTR) 136. Each transmitter unit 136 processes (e.g., converts to analog, amplifies, filters, and frequency upconverts) its output chips and generates a modulated signal. T modulated signals from T transmitter units 136a through 136t are transmitted from T antennas 138a through 138t, respectively.
[0024] At receiver 150, multiple (R) antennas 152a through 152r receive the trarismitted signals via various signal paths and provide R received signals to R receiver units (RCVR) 154a through 154r, respectively. Each receiver unit 154 processes (e.g., filters, amplifies, frequency downconverts, and digitizes) its received signal and provides samples to a channel and noise estimator 160 and a space-time equalizer 162. Estimator 160 derives channel and noise estimates based on the samples, as described below. Space-time equalizer 162 derives weights based on the channel and noise estimates, performs equalization on the samples with the weights, and provides data chip estimates to T CDMA demodulators (Demod) 170a through 1701 Each CDMA demodulator 170 processes its data chip estimates in a manner complementary to the
processing by CDMA modulator 130 and provides data symbol estimates. A receive (RX) data processor 180 processes (e.g., symbol demaps, dcintcrlcavcs, and decodes) the data symbol estimates and provides decoded data. In general, the processing by CDMA demodulator 170 and RX data processor 180 is complementary to the processing by CDMA modulator 130 and TX data processor 120, respectively, at transmitter 110.
[0625] Controllers/processors 140 and 190 direct operation of various processing
units at transmitter 110 and receiver 150, respectively. MIMOries 142 and 192 store
data and program codes for transmitter 110 and receiver 150, respectively.
[0026] FIG. 2 shows a block diagram of a CDMA modulator 130 for one transmit antenna. CDMA modulator 130 may be used for each of CDMA modulators 130a through 330t in FIG. 1. CDMA modulator 130 includes a traffic channel processor 210 for each traffic channel used for traffic data and a pilot channel processor 220 for pilot. Within processor 210 for traffic channel m, a spreader 212 spreads data symbols with an orthogonal code cm(k) for traffic channel m. A multiplier 214 scales the output of
spreader 212 with a gain gUm and provides data chips x,m (k) for traffic channel m on
transmit antenna t. Within pilot channel processor 220, a spreader 222 spreads pilot symbols with an orthogonal code clp(k) for pilot. A multiplier 224 scales the output of
spreader 222 with a gain gtJ> and provides pilot chips xtp (k) for the pilot channel on
transmit antenna t. A summer 230 sums the chips for all traffic and pilot channcLs. A scrambler 232 multiplies the output of summer 230 with a scrambling sequence s(k} for transmitter 110 and provides output chips xt(k) for transmit antenna t.
[0027] The orthogonal codes for the traffic and pilot channels may be orthogonal variable spreading factor (OVSF) codes used in W-CDMA, Walsh codes used in cdma2000, and so on. In an embodiment, T different orthogonal codes are used for the pilot for the T transmit antennas to allow receiver 150 to estimate the channel response for each transmit antenna. The remaining orthogonal codes may be used for each of the T transmit antennas. For this embodiment, the pilot orthogonal codes for the T transmit antennas are distinct whereas the traffic orthogonal codes may be reused for the T transmit antennas. The spreading and scrambling may also be performed in other manners than the manner shown in FIG. 2. [0028] The output chips for each transmit antenna t may be expressed as:
(Equation Removed) (l)
where x, (fc) is a Kxl vector with K. output chips sent from transmit antenna t, xt (k) is the output chip sent in chip period k from transmit antenna t, crx is the gain for the output chips, and " r" denotes a transpose.
K may be set as K = E + L —1, where E is the span of space-time equalizer 162 at receiver 150 and L is the delay spread of the MIMO channel between transmitter 110 and receiver 150. Vector x,(£) would then contain E + L-l chips to be operated on by the space-tune equalizer at the receiver.
[0029] In an embodiment, the receiver digitizes the received signal from each receive antenna at V times the chip rate for an oversampling factor of V and obtains V samples for each chip period. The samples for each receive antenna r may be expressed as:
(Equation Removed) Eq(2)
where yr(k) is a V - E x l vector of samples for receive antenna r, and yr,v (k) is the v_th sample in chip period k for receive antenna r. [0030] The noise observed by the samples in y (k) may be expressed as:
(Equation Removed) Eq(3)
where nr(_k) is a V- E x 1 noise vector for receive antenna r, and ne,y(k) is the noise observed by yr „(k) .
[0031] nr(k) includes channel noise, receiver noise, and interference from other transmitters, which are collectively referred to as "background" noise. As used herein, the term "noise" may refer to just background noise or background noise and inter-strcam interference from other transmit antennas. For example, channel and noise estimation may refer to channel estimation and either background noise estimation or background noise and interference estimation. [0032] The samples for all R receive antennas may be expressed as:

(Equation Removed)Eq(4)

where Hr, is a V-ExK muitipath chaimel matrix between transmit antenna t and Teceive antenna r. Hr,, may be given as:
(Equation Removed)
where hr,i,v(l) is a complex channel gain between transmit antenna r and receive
antenna r for the l-th channel tap at the v-th sampling time instant.
[0033] The impulse response between transmit antenna r and receive antenna r has a length of L chips and may be oversampled at V times chip rate to obtain V • L channel gains. These V-L channel gains may be arranged into V rows such that each row contains L channel gains for one sampling time instant v. Sample yr ,v(k) for sampling
time instant v is obtained by convolving the transmitted data chips in x/Ar) with a row of channel gains for that sampling time instant. Hr, contains V - E rows for the V - E samples in y (k). Each row of Hr,, represents the channel impulse response for one sample in y (k). [0034] Equation (4) may be rewritten as follows:
(Equation Removed)Eq(6)
where y{k)=[y_T(k) y2T(k) ... YRT(k)]T is an R-V-Exl overall received sample vector for all R receive antennas, x(k) = [X1T (k) X2T(k) ... XTT (k)]T is a T - K x 1 overall transmitted chip vector for all T transmit antennas,
n(k) = [nT1(k) nT2(k) ... DTR(k)]T is an R-V-Exl overall noise vector for all R
receive antennas, and H is an R-V-ExT-K overall channel response matrix.
[0035] The overall channel response matrix H may be given as:
(Equation Removed)Eq(7)
The first K columns of H are for transmit antenna 1, the next K colximns of H are for
transmit antenna 2, and so on, and the last K columns of H are for transmit antenna T.
Each column of H is denoted as h,- and contains the channel gains observed by one
transmitted chip x, (k) .
[0036] The transmitted chips may be recovered by space-time equalizer 162, as
follows:
(Equation Removed) Eq(8)
where w, is an R - V - E x 1 equalizer weight vector for transmit antenna t,
x,(k + D) is an estimate of data chip x,(k 4- D) sent from transmit antenna t, and " B " denotes a conjugate transpose.
[0037] For each chip period k, vector y(&) contains samples for all R receive
antennas for the E most recent chip periods. These samples contain components of K = E + L -1 most recent data chips sent from each transmit antenna. Vector y(k) is
multiplied with wf to obtain an estimate of one data chip from one transmit antenna. T weight vectors Wj through wT may be formed for the T transmit antennas and used to obtain estimates of T data chips sent from the T transmit antennas in one chip period. [0038] In equation (8), D denotes the delay of the space-time equalizer. For each chip period k, the space-time equalizer provides an estimate of data chip xt (k + D) sent
D chip periods later, instead of an estimate of data chip xr(k) sent in the k-th chip
period. D is determined "by the design of the equalizer and, in general, 1≤D≤(E + L-1).
[0039] The weight vectors w, for the space-time equalizer may be derived based on
various receiver processing techniques such as a ininimum mean square error (MMSE) technique, a zero-forcing (ZF) technique, a maximal ratio combining (MRC) technique, and so on. Weight matrices may be derived based on the MMSE, ZF, and MRC techniques, as follows:
(Equation Removed)Eq (9)Eq (10)Eq(ll)
where R„ is a noise covariance matrix and are T . K x R • V - E
weight matrices for the MMSE, ZF and MRC techniques, respectively. The weight
vectors wf may be obtained from the rows of W_r, WfJ- and Wf„
[0040] The noise covariance matrix R„ may be expressed as:
(Equation Removed) Eq (12)
where E{ } denotes an expectation operation. R„ is indicative of the background noise
statistics and has a dimension of R-V-ExR- V-E.
[0041] The channel and noise estimation techniques described herein may be used
with various receiver processing techniques. For clarity, the following description is for
the MMSE technique.
[0042] For the MMSE techndque, the weight vectors w,, for t = l,...,T, may be
expressed as:
(Equation Removed)
where ht,D = h(t-1)K+D is a channel response vector for transmit antenna t at delay D, Ryy is a matrix indicative of the signal, noise and interference statistics, and I is a set that includes 1 through T • K but excludes (t -1) • K + D .
[0043] As shown in equation (13), Ryy includes H-HH and R„. H-HH includes the desired signal as well as interference from other transmit antennas. R„ includes the background noise. Hence, Ryy,, includes the desired signal, the background noise, and the interference from other transmit antennas. For the last equality in equation (13), each summation is over the outer products of all columns of H except for the column for the desired signal, which is htD . hlyD mcludes on-time components arriving at the
receiver at delay D. These on-time components include (1) on-time signal component from the desired transmit antenna t and (2) on-time inter-stream interference from the other T —1 transmit antennas also arriving at delay D. Equation (13) minimizes E { wH-y(k)-xt(k + D)\2 }, which is the mean square error between the transmitted
data chip x,(k + D) and its estimate wf -y(k).
[0044] Enhanced weight vectors w, that take into account the despreading effect of
the subsequent CDMA demodulator may be expressed as:
(Equation Removed)Eq(14)
where G is a gain for a traffic channel being received from transmit antenna t,
■ It is a set that includes 1 through T but excludes t, and
Q is a set that includes 1 through T • K but excludes (i -1) • K + D, for i = 1,..., T.
[0045] For simplicity, equation (14) assumes the use of the same gain G for all transmit antennas and all traffic channels/orthogonal codes/streams. If J orthogonal codes are used for each transmit antenna and if each orthogonal code has a length of SF,
then gain G may be given as G = SF/J . In general, different gains may be used for different orthogonal codes and/or different transmit antennas.
[0046] As shown in equations (13) and (14), the MMSE weight vectors may be derived based on the channel response vectors h, through hT.K and the noise covariance matrix £» • The channel response vectors may be derived based on a pilot, which may be transmitted using code division multiplexing (CDIvf), time division multiplexing (TDM), frequency division multiplexing (FDM), and so on. For clarity, the following description assumes the use of a CDM pilot transmitted as shown in FIG. 2. The channel response vectors and the noise covariance matrix may be estimated directly or indirectly with various schemes, as described below.
[0047] In a first channel and noise estimation scheme, the channel and noise estimates are derived directly from the received samples. For this scheme, the channel response vectors may be estimated as follows:
(Equation Removed)Eq(lS)
where p, (k) is a pilot chip sent in chip period k from transmit antenna /, Np is the length of the orthogonal code for the pilot,
Eq, is the energy per pilot chip,
LPFchuuiei denotes an, averaging operation for the channel estimate, and
htp is an estimate of σx .ht,D.
[0048] The pilot chips for transmit antenna t may be given as:
(Equation Removed) Eq(16)
where d(j) is a pilot symbol sent in pilot symbol period./,
ctp (k) is the orthogonal code for the pilot for transmit antenna tz and s(k) is the scrambling code for the transmitter.
[0049] In equation (15), the overall sample vector y(k) is multiplied with the
complex conjugated pilot chip pt(k + D) for delay D and accumulated over the length
of the pilot orthogonal code to obtain an initial estimate of ht,D. The initial estimate
may be filtered over multiple pilot symbols to obtain ht,D, which is an improved
estimate of ht,D. The filtering may be performed based on a finite impulse response
(FIR) filter, an infinite impulse response (IIR) filter, and so on. For example, a single-tap ICR. filter with a time constant of 0.33, 0.2, or some other value per slot may be used for the channel filter, LPF Chsnnel. For W-CDMA, each slot spans 2560 chips and covers 10 pilot symbols sent with a 256-chip OVSF code. For cdma2000, each slot spans 768 chips and may cover 12 pilot symbols sent with a 64-chip Walsh code. [0050] The signal, noise and interference statistics may be estimated as follows:
(Equation Removed) Eq (17)
where LPFdoise denotes an averaging operation for Ryy, and
Ryy, is an estimate of Jg.Ryy .
The noise filter, LPFnoise, may be the same or different from the channel filter, LPFchannel. [0051] The weight vectors may then be derived as follows:
(Equation Removed)Eq(18)
Equation (18) is equivalent to equation (13) if ht,D and Ryy are accurate estimates of σx.hD and Ryy, respectively. The SNR of the data chip estimates from the space-time equalizer may be expressed as:
(Equation Removed) Eq (19)
[0052] The first scheme derives a single estimate for Ryy,. This estimate, Ryy ,
includes components for both summations in the first equality for equation (14). The two components are not accessible to apply the gain G for the desired traffic channel, and the enhanced weight vectors w, are not derived from Ryy.
[0053] FIG. 3 shows a block diagram of a channel and noise estimator 160a, which implements the first channel and noise estimation scheme and is an embodiment of channel and noise estimator 160 in FIG. 1. Estimator 160a includes a serial-to-parallel (S/P) converter 310, a channel estimator 320, and a signal, noise and interference statistics estimator 330. S/P converter 310 receives the samples from R receiver units 154 a through 154r and forms vector y(k).
[0054] Within channel estimator 320, a pilot correlator 322 correlates y(&) with a pilot sequence p,(k) for each transmit antenna ./ and provides initial channel estimates for all transmit antennas. A channel filter 324 filters the initial channel estimates and provides final channel estimates, ht,d. Pilot correlator 322 and filter 324 perform processing for equation (15). Within estimator 330, a unit 332 computes the outer product of y(A). A noise filter 334 then filters the output of unit 332 and provides the
estimated signal, noise and interference statistics, Ryy • Unit 332 and filter 334 perform
processing for equation (17). The scaling factor in equation (15) may be accounted for in various units, e.g., in filter 324.
[0055] FIG. 4 shows an embodiment of channel estimator 320 in FIG. 3. For this embodiment, pilot correlator 322 in FIG. 3 is implemented with R pilot correlators 322a through 322r and channel filter 324 is implemented with R channel filters 324a through 324r. One set of pilot correlator 322r and channel filter 324r is provided for each receive antenna.
[0056] The samples yhv(Jc) through yR,V(A) from the R receive antennas are provided to R pilot correlators 322a through 322r, respectively. Within each pilot correlator 322r, the samples vr,v(fc) are provided to V-E delay elements 410 that are
coupled in scries. Each delay clement 410 provides a delay of one sample period. The V • E delay elements 410 provide their delayed samples to V • E multipliers 412. Each multiplier 412 multiplies its delayed sample with the complex conjugated pilot chip pi(k + D) for delay D. V-E accumulators 414 couple to the V-E multipliers 412.
, Each accumulator 414 accumulates the output of an associated multiplier 412 over the length of the pilot orthogonal code, or Np chips, and provides an initial channel gain estimate h't,r,v(l) for each pilot symbol period.
[0057] Within each channel filter 324r, V • E channel filters (LPFs) 420 couple to
the V • E accumulators 414 in an associated pilot correlator 322r. Each filter 420 filters the initial channel gain estimates from an associated accumulator 414 and provides a final channel gain estimate h,t,r,v(l) for each update interval, e.g., each pilot symbol period. An S/P converter 430 receives the final channel gain estimates from filters 420 for all R receive antennas and provides a channel response vector ht D for each update
interval.
[0058] In general, the processing units shown in FIGS. 3 and 4 may be implemented in various manners. For example, these units may be implemented with dedicated hardware, a shared digital signal processor (DSP), and so on.
[005.9] FIG. 5 shows an embodiment of a process 500 for performing channel and
noise estimation based on the first scheme. Samples are obtained from multiple receive antennas for a MIMO transmission sent from multiple transmit antennas (block 512). Channel estimates are derived, e.g., by correlating the samples with at least one pilot sequence (block 514). The channel estimates may comprise T channel response vectors h,t,d for t — 1,..., T , for T transmit antennas at delay D. The channel response vector for each transmit antenna may be obtained based on a pilot sequence for that transmit antenna. The signal, noise and interference statistics are estimated based on the samples, e.g., by computing cross product of the samples and filtering the cross product results (block 516). The channel estimates may be derived with a first filter having a first bandwidth. The estimated signal, noise, and interference statistics may be derived with a second filter having a second bandwidth. The first and second filter bandwidths
may be the same or different and may be fixed or configurable, e.g., adjusted based on
channel conditions. Equalizer weights arc derived based on the channel estimates and
the estimated signal, noise and interference statistics (block 518). The samples are
filtered with the equalizer weights to obtain estimates of the data chips sent from the
transmit antennas (block 520).
[0060] In a second channel and noise estimation scheme, the noise estimation is
performed based on an estimate of the total received energy, Tσ, at the receiver and an
estimate of the signal and interference energy, lor- The terms "energy" and "power" are
often used interchangeably.
[0061] For the second scheme, the elements of vectors σx h1 through σx.hT.K
may be estimated as follows:

(Equation Removed)Eq (20)

where Ar,t,v(l) is an estimate of channel gain hr,t,v{l). Channel gain estimates may be obtained for r = l, ..., R, r = l, ..., T, v = l, ..., V, and / = 0, ..., L-l. Equation (20) is equivalent to equation (15). However, hr,t,v(l) may be computed for different values of r, t, v and / with equation (20) whereas h,D may be computed for different values of t and a specific value of D with equation (15). The channel gain estimates hr,t,v{l) may
be used to form h,, which is an estimate of h,, for i = 1, ..., T • K
[0062] For the second scheme, the background noise may be assumed to be spatio-temporally white, so that Rn = σn2.I,, where σ2n is the noise variance and I is an identity matrix. The noise variance may be estimated as follows:
(Equation Removed) Eq(21)
where I is the estimated total received energy, and
Ior is the estimated signal and interference energy. [0063J In an embodiment, It and Iar may be derived as follows:

(Equation Removed)
[0064] In another embodiment, Jor may be derived by stmiming channel gains with sufficient energy. A channel gain may be considered to be sufficiently strong if, e.g., hr,t,v(l)2≥Eth, where Eth is a threshold. Eth may be a fixed value or a configurable
value that may be derived based on the total energy for all channel gains. [0065] The noise covariance matrix may then be estimated as follows:
(Equation Removed)Eq(24)
where R„ is an estimate of Rn. Rn and h, may be used to derive w, as shown in equation (13) or w, as shown in equation (14).
[0066] FIG. 6 shows a block diagram of a channel and noise estimator 160b, which implements the second channel and noise estimation scheme and is another embodiment of channel and noise estimator 160 in FIG. 1. Estimator 160b includes a multiplexer (Mux) 610, a channel estimator 620, and a noise estimator 630. Multiplexer 610 receives the samples from R receiver units 154a through 154r and provides a stream of samples yr,v(k) in a desired order.
[0067] Within channel estimator 620, a pilot correlator 622 correlates yr v(k) with a
pilot sequence pt(k) for each transmit antenna and provides initial channel gain
estimates for all transmit antennas. A channel filter 624 filters the initial channel gain
estimates and provides final channel gain estimates, Ar,t,v(/). Pilot correlator 622 and
filter 624 perform the processing shown in equation (20). An S/P converter 626 receives the final channel gain estimates for all transmit antennas and provides channel
response vectors h,-, for / = 1, ..., T • K.
[0068] Noise estimator 630 estimates the total received energy I0, the signal and
interference energy Ior, and the noise For the I0 estimate, an energy computation
unit 642 computes the energy of each sample. An accumulator 644
accumulates the sample energies across the V sample periods and the R receive antennas and provides an initial I0 estimate for each chip period. A noise filter 646
filters the initial I0 estimates and provides the final I0 estimate, I0. For the Ior estimate,
an energy computation unit 652 computes the energy of each channel tap as | kt (/) |2.
An accumulator 654 accumulates the channel tap energies across the V sample periods, the L channel taps, the T transmit antennas, and the R receive antennas and provides an initial Ior estimate. The accumulation may also be performed over channel taps with sufficient energy. A noise filter 656 filters the initial /„. estimates and provides the final
Ior estimate, Ior. For the σ2 estimate, a summer 658 subtracts Iar from /„ and
provides the noise variance estimate, Σ2 . A unit 660 forms a noise covariance matrix
R, based on ΣN„2 , as shown in equation (24). The scaling factors in equations (20), (22)
and (23) maybe accounted for in various units, e.g., in filters 624, 646 and 656. [0069] FIG. 7 shows an embodiment of a process 700 for performing channel and noise estimation based on the second scheme. Samples are obtained from multiple receive antennas for a MIMO transmission sent from multiple transmit antennas (block 712). Channel estimates are derived, e.g., by correlating the samples with at least one pilot sequence (block 714). The total received energy I0 is estimated based, on the samples, e.g., by computing the energy of each sample, accumulating the sample energies across the receive antennas and for a given time interval, and filtering the accumulated results for different time intervals (block 716). The signal and interference energy Ior is estimated based on the samples, e.g., by computing the energy of each channel tap in the channel estimate, accumulating the energies of all or sufficiently strong channel taps for a given time interval, and filtering the accumulated results for different time intervals (block 718). The channel estimates may be derived with a first filter having a first bandwidth. The total received energy as well as the signal and interference energy may be estimated with a second filter having a second bandwidth that may be the same or different from the first bandwidth. The noise σn2 Ls estimated based on the estimated total received energy and the estimated signal and interference energy (block 720). Equalizer weights are derived based on the channel estimates and the estimated noise (block 722). The samples are filtered with the equalizer weights to obtain estimates of the data chips sent from the transmit antennas (block 724).
[0070] The second scheme can provide good performance for operating scenarios in which the background noise is spatio-tcmporally white. The second scheme also works well for a severe multipath environment since (a) the background noise Rn is less significant and (b) the dominant component of Ryy is likely to be multipath interference
that may be accurately estimated based on the channel estimates obtained from equation (20).
[0071] In a third channel and noise estimation scheme, Ryy is decomposed into on-time components and remaining components, which are estimated separately. For the third scheme, initial channel estimates may be derived as follows:
(Equation Removed) Eq (25)
where hr,d (j) is an estimate of hr,d for pilot symbol period j.
[0072] The channel response vectors may then be estimated as follows:
(Equation Removed)Eq (26)
Equations (25) and (26) are equivalent to equation (15).
[0073] Equations (13) and (14) for the equalizer weights may be rewritten as
follows:
(Equation Removed)
where Rat is a matrix indicative of signal and on-time interference statistics, and
R,. is a matrix indicative of noise and multipath interference statistics.
[0074] Rot may be expressed as:
(Equation Removed)
Rot includes the desired signal as well as the on-time interference from the other
transmit antennas.
[0075] Rc may be expressed as:
(Equation Removed)
where g is a set that includes 1 through TK but excludes (/-1)K + D, for i = 1,..., T. Rc includes (1) multipath interference, which is interference that is not on-time and represented by the summation in equation (30), and (2) background noise, which is represented by Rn.. Rc,. does not include contribution from the signal and on-time interference, which is represented by ht,D. Hence, Rc may be estimated by performing differential-based multipath plus noise estimation. [0076] To estimate Rc, ht,D(j) is first differentiated as follows:
(Equation Removed) Eq(31)
The differential operation in equation (31) cancels the on-time components in ht,D(j) if the channel response remains constant over two consecutive pilot symbol periods. A
scaling factor of 1 / is introduced in equation (31) so that the differential operation does not change the second-order moment of the multipath interference and the background noise. An estimate of Rc may then be derived as follows:
(Equation Removed)
Rc docs not include contributions from the on-time components, which arc removed by
the differential operation, in equation (31).
[0077] An estimate of Rot may be derived as follows:
(Equation Removed)
Eq(33)
[0078] For the weight vectors shown in equation (13) and (27), Rc may be augmented with Rot to obtain an estimate of Ryy,, as follows:
(Equation Removed) Eq(34)
The weight vectors may then be derived as:
(Equation Removed) Eq(35)
[0079] For the enhanced weight vectors shown in equation (14) and (28), Rc may be augmented with Rot, to obtain an estimate of Ryy, as follows:
(Equation Removed) Eq (36)
The enhanced weight vectors may then be derived as:
(Equation Removed) Eq(37)
[0080] FIG. 8 shows a block diagram of a channel and noise estimator 160c, which
implements the third channel and noise estimation scheme and is yet another embodiment of channel and noise estimator 160 in FIG. 1. Estimator 160c includes an S/P converter 810, a channel estimator 820, and a signal, noise and interference statistics estimator 830. S/P converter 810 receives the samples from R receiver units 154a through 154r and forms vector y(k).
[0081] Within channel estimator 820, a pilot correlator 822 correlates y(k) with a
pilot sequence ps(k) for cach transmit antenna t and provides an initial channel
estimate, ht,D(j) for each transmit antenna in each pilot symbol period J. A channel
filter 824 filters the initial channel estimates and provides final channel estimates, hc,D .
Pilot correlator 822 performs the processing shown in equation (25). Filter 824 performs the processing shown in equation (26).
[0082] Within estimator 830, a unit 832 receives and differentiates the initial channel estimates, ht,D (J). as shown in equation (31) and provides the difference, Δht,D(j). A unit 834 computes the outer product of Δht,D(j) • -A- unit 836 sums the output of unit 834 across all T transmit antennas. A noise filter 838 filters the output from unit 836 and provides the estimated noise and raultipath interference statistics, Rc. Units 834, 836 and 838 perform the processing shown in equation (32). A unit 844 computes the outer product of ht,D . A unit 846 sums the output of unit 844 across all T transmit antennas and provides the estimated signal and on-time interference statistics, Rot . Units 844 and 846 perform the processing shown in equation (33). A matrix summer 850 sums Rc. with Rot, and provides the estimated signal, noise and interference statistics, Ryy. The scaling factors in equations (26), (31) and (32) may be
accounted for in various units, e.g., in filters 824 and 838.
[0083] Rc. and Rot may also be estimated in other manners. For example, an
estimate of Rot, may be obtained by computing the outer product of ht,D(J), surorning across T transmit antennas, and filtering the summed results. As another example, an estimate of Rc may be obtained by filtering Δht,d(J), computing the outer product of the filtered Δht,d(/) , and summing across the T transmit antennas.
[0084] FIG. 9 shows an embodiment of a process 900 for performing channel and
noise estimation based on the third scheme. Samples are obtained from multiple receive antennas for a MIMO transmission sent from multiple transmit antennas (block 912). First or initial channel estimates may be derived, e.g., by correlating the samples with at least one pilot sequence (block 914). The first channel estimates are filtered with a first filter to obtain second or final channel estimates (block 916).
[0085] The signal and on-time interference statistics are estimated based on the samples, e.g., by computing the cross-product of the second channel estimates and summing across the transmit antennas (block 918). The noise and multipath interference statistics are also estimated based on the samples, e.g., by differentiating the first channel estimates, computing cross-product of the differentiated results, and filtering the cross-product results with a second filter (block 920). The first and second filters may have the same or different bandwidths, which may be fixed or configurable, e.g., adjusted based on channel conditions. The signal, noise and interference statistics
are then estimated based on the estimated signal and on-time interference statistics and
the estimated noise and multipath mtcrfcrcncc statistics (block 922).
[0086] Equalizer weights are derived based on the second channel estimates and the
estimated signal, noise and interference statistics (block 924). The samples are filtered
with the equalizer weights to obtain estimates of the data chips sent from the transmit
antennas (block 926).
[0087] As shown in equation (26), the on-time components at delay D may be
averaged by the channel LPF channel, to obtain ht,D, which is used to derive Rol.
The background noise and multipath interference may be averaged by the noise filter,
LPFnoisc, to obtain Rc. The use of two filters to average the desired signal in hjD and
the noise and multipath interference in Rc may be beneficial in some environments. For example, in a single-path high-velocity channel, a small bandwidth or large time constant may be used for the noise filter to improve the accuracy of Rc, and a large bandwidth or small time constant may be used for the channel filter to prevent utilization of stale or obsolete channel estimates h, D. The noise and multipath
interference estimation is typically more accurate with a smaller bandwidth for the noise filter. A smaller noise filter bandwidth is beneficial for a severe multipath channel in order to sufficiently suppress large uncorrelated components associated with multipath. However, the bandwidth of the noise filter should not be reduced too much in a severe multipath high-velocity channel since multipath component becomes the dominant component that determines Rc, and the multipath component may become outdated if the noise filter bandwidth is too small.
[0088] The third scheme provides good performance for many channel environments, especially for a single-path high geometry channel. The use of the robust diffcrcntial-bascd noise estimation allows,the third scheme to work well in various channel conditions. The third scheme also has lower computational complexity than the first scheme since the outer products in equations (32) and (33) are performed every pilot symbol period instead of every chip period. The third scheme also provides good performance even with a colored background noise.
[0089] For the third scheme, the differential operation in equation (31) assumes that
the channel is constant over two consecutive pilot symbol periods. A sufficiently short pilot symbol duration may be used for a high-velocity channel so that the noise and
multipath interference estimation is not outdated. Computer simulation results show that a pilot symbol duration of 256 chips or less for HSDPA in W-CDMA provides good performance with 30km/h velocity.
[0090] In general, the estimation of the multipath interference in Rc may be improved with more vectors for h't,DO)- The number of h't,D(j) vectors may be increased by taking partial integration of the pilot symbols. For example, if the pilot symbol duration is 512 chips long, then the integration length may be set to 128 chips, and four ht,D (j) vectors may be obtained per pilot symbol. The partial integration may result in crosstalk between orthogonal pilots from different transmit antennas. However, since the orthogonal pilot patterns and the channel estimates h,t,D are known, the pilot crosstalk may be estimated and subtracted out. This approach may still suffer from the leakage (crosstalk) of some OCNS channels if they have a symbol length greater than 128, but it can be acceptable if the power of the corresponding OCNS is small. Alternatively, in order to maximize the performance, we may take the length of pilot symbols shorter than that of some OCNS symbols in the system design. [0091] In a fourth channel and noise estimation scheme, multiple estimation schemes are supported, and one estimation scheme is selected for use based on channel conditions. In an embodiment, the second scheme is selected for a severe multi-path high-velocity channel and the third scheme is selected for a single-path high-geometry channel. Other combinations of channel and noise estimation schemes may also be used. [0092] The channel conditions may be detected in various manners and with various metrics. In an embodiment, the channel conditions are characterized by delay spread. For this embodiment, a searcher may search for signal paths or multipaths, e.g., similar to the search performed for a CDMA receiver. The delay spread may be computed as the difference between the earliest and latest arriving signal paths at the receiver. In another embodiment, the channel conditions are characterized by a path energy ratio. For this embodiment, the energy of the strongest signal path is computed, and the combined energy of all other signal paths identified as significant (e.g., with energy exceeding a threshold Eth) is also computed. The path energy ratio is the ratio of the combined energy for the significant paths to the energy for the strongest path. A low padi energy ratio may indicate a single path environment whereas a high path energy ratio may indicate a multipath environment. In yet another embodiment, the velocity of
the receiver is estimated, e.g., based on an estimate of Doppler and/or some other velocity-indicating parameter.
[0093] In an embodiment, the third estimation scheme is selected if the delay spread is smaller than a delay threshold Dth or if the path energy ratio is less than a predetermined threshold Pa, or if the velocity is below a velocity threshold Vth. The second estimation scheme may be selected otherwise. If the velocity is low, then the bandwidth of the noise filter may be reduced to improve noise estimation. In general, the third estimation scheme may be used if a sufficient number of noise and multipath interference samples that are relatively current can be obtained, e.g.; via a short pilot symbol period or small bandwidth for the noise filter. Otherwise, the second estimation scheme may be used.
[0094] FIG. 10 shows an embodiment of a process 1000 for performing channel
and noise estimation based on the fourth scheme. Samples are obtained from multiple receive antennas for a MIMO transmission sent from multiple transmit antennas (block 1012). Channel conditions arc detennined based on the samples (block 1014). The channel conditions may be characterized by (1) delay spread, which may be estimated based on the earliest and latest arriving signal paths, (2) single path or multipath environment, which may be determined based on the energies of the signal paths detected at the receiver, (3) velocity, which may be estimated based on Doppler, and/or (4) possibly other criteria.
10095] A channel and noise estimation scheme is selected from among multiple
channel and noise estimation schemes based on the channel conditions (block 1016). The multiple channel and noise estimation schemes may include all or any combination of the schemes described above. For example, the second and third channel and noise estimation schemes described above may be supported. The third channel and noise estimation scheme may be selected if the channel conditions indicate a small delay spread, a single path environment, low velocity, or a combination thereof. The second channel and noise estimation scheme may be selected if the channel conditions indicate a large delay spread, a multipath environment, high velocity, or a combination thereof. Channel and noise estimation is then performed based on the selected channel and noise estimation scheme (block 1018).
[0096] The channel and noise estimation techniques described herein may be implemented by various means. For example, these techniques may be implemented in hardware, firmware, software, or a combination thereof. For a hardware
implementation, the processing units used to perform channel and noise estimation may be implemented within one or more application specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field programmable gate arrays (FPGAs), processors, controllers, micro-controllers, microprocessors, electronic devices, other electronic units designed to perform the functions described herein, or a combination thereof.
[0097] For a firmware and/or software implementation, the techniques may be implemented with modules (e.g., procedures, functions, and so on) that perform the functions described herein. The firmware and/or software codes may be stored in a MIMOry (e.g., MIMOry 192 in FIG. 1) and executed by a processor (e.g., processor 190). The MIMOry may be implemented within the processor or external to the processor, in which case it can be communicatively coupled to the processor via various means as is known in the art.
[0098] 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.

CLAIMS
1. An apparatus comprising:
at least one processor to obtain samples from multiple receive antennas for a multiple-input multiple-output (MEMO) transmission sent from multiple transmit antennas, to derive channel estimates by correlating the samples with at least one pilot sequence, and to estimate signal, noise and interference statistics based on the samples; and
a memory coupled to the at least one processor.
2. The apparatus of claim 1, wherein the at least one processor derives the
channel estimates with a first filter having a first bandwidth, and estimates the signal,
noise and interference statistics with a second filter having a second bandwidth.
3. The apparatus of claim 1, wherein the at least one processor derives equalizer "weights based on the channel estimates and the estimated signal, noise and mterference statistics, and filters the samples with the equalizer weights.
4. The apparatus of claim 1, wherein multiple signals are transmitted from the multiple transmit antennas, and wherein each signal comprises multiple data streams multiplexed with different orthogonal codes.
5. A method comprising:
obtaining samples from multiple receive antennas for a multiple-input multiple-output (MIMO) transmission sent from multiple transmit antennas;
deriving channel estimates by correlating the samples with at least one pilot sequence; and
estimating signal, noise and interference statistics based on the samples.
6. The method of claim 5, wherein the deriving the channel estimates
comprises deriving the channel estimates with a first filter having a first bandwidth, and
wherein the estimating the signal, noise and interference statistics comprises estimating
the signal, noise and interference statistics with a second filter having a second
bandwidth.
7. The method of claim 5, further comprising:
deriving equalizer weights based on the channel estimates and the estimated signal, noise and interference statistics; and
filtering the samples with the equalizer weights.
8. An apparatus comprising:
means for obtaining samples from multiple receive antennas for a multiple-input multiple-output (MIMO) transmission sent from multiple transmit antennas;
means for deriving channel estimates by correlating the samples with at least one pilot sequence; and
means for estimating signal, noise and interference statistics based on the samples.
9. The apparatus of claim 8, wherein the means for deriving the channel
estimates comprises means for deriving the channel estimates with a first filter having a
first bandwidth, and wherein the means for estimating the signal, noise and interference
statistics comprises means for estimating the signal, noise and interference statistics
with a second filter having a second bandwidth.
10. The apparatus of claim 8, further comprising:
means for deriving equalizer weights based on the channel estimates and the estimated signal, noise and interference statistics; and
means for filtering the samples with the equalizer weights.
11. An apparatus comprising:
at least one processor to obtain samples from multiple receive antennas for a multiple-input multiple-output (MIMO) transmission sent from multiple transmit antennas, to estimate total received energy based on the samples, to estimate signal and interference energy based on the samples, and to estimate noise based on the estimated total received energy and the estimated signal and interference energy; and
a memory coupled to the at least one processor.
12. The apparatus of claim 11, wherein the at least one processor derives
channel estimates by correlating the samples with at least one pilot sequence, and
estimates the signal and interference energy based on the channel estimates.
13. The apparatus of claim 12, wherein the at least one processor identifies channel taps with sufficient strength from among all channel taps for the channel estimates, and determines the signal and interference energy based on the identified channel taps.
14. The apparatus of claim 11, wherein the at least one processor computes energy of each of the samples, accumulates energies of the samples across the multiple receive antennas, and filters accumulated results for different time intervals to obtain the estimated total received energy.
15. The apparatus of claim 12, wherein the at least one processor computes energy of each of multiple channel taps for the channel estimates, accumulates energies of the multiple channel taps, and filters accumulated results for different time intervals to obtain the estimated signal and interference energy.

16. The apparatus of claim 12, wherein the at least one processor derives the channel estimates with a first filter having a first bandwidth, and estimates the total received energy and the signal and interference energy with a second filter having a second bandwidth.
17. The apparatus of claim 16, wherein the at least one processor selects the first and second bandwidths based on channel conditions.
18. A method comprising:
obtaining samples from multiple receive antennas for a multiple-input multiple-output (MEMO) transmission sent from multiple transmit antennas;
estimating total received energy based on the samples;
estimating signal and interference energy based on the samples; and
estimating noise based on the estimated total received energy and the estimated signal and interference energy.

19. The method of claim 18, wherein the estimating the total received energy
comprises
computing energy of each of the samples, and accumulating energies of the samples.
20. The method of claim 18, wherein the estimating the signal and
interference energy comprises
deriving channel estimates comprising multiple channel taps, computing energy of each of the multiple channel taps, and accumulating energies of the multiple channel taps.
21. An apparatus comprising:
means for obtaining samples from multiple receive antennas for a multiple-input multiple-output (MTMO) transmission sent from multiple transmit antennas;
means for estimating total received energy based on the samples;
means for estimating signal and interference energy based on the samples; and
means for estimating noise based on the estimated total received energy and the estimated signal and interference energy.
22. The apparatus of claim 21, wherein the means for estimating the total
received energy comprises
means for computing energy of each of the samples, and means for accumulating energies of the samples.
23. The apparatus of claim 21, wherein the means for estimating the signal
and interference energy comprises
means for deriving channel estimates comprising multiple channel taps, means for computing energy of each of the multiple channel taps, and means for accumulating energies of the multiple channel taps.
24. An apparatus comprising:
at least one processor to obtain samples from multiple receive antennas for a multiple-input multiple-output (MEMO) transmission sent from multiple transmit
antennas, to estimate signal and on-time interference statistics based on the samples, to estimate noise and multipath interference statistics based on the samples, and to estimate signal, noise and interference statistics based on the estimated signal and on-time interference statistics and the estimated noise and multipath interference statistics; and
a memory coupled to the at least one processor.
25. The apparatus of claim 24, wherein the at least one processor derives channel estimates by correlating the samples with at least one pilot sequence.
26. The apparatus of claim 24, wherein the at least one processor derives first channel estimates based on the samples, and filters the first channel estimates to obtain second channel estimates.
27. The apparatus of claim 26, wherein the at least one processor differentiates the first channel estimates to obtain differentiated results, and estimates the noise and multipath interference statistics based on the differentiated results.
28. The apparatus of claim 26, wherein the at least one processor estimates the signal and cm-time interference statistics based on the second channel estimates.
29. The apparatus of claim 24, wherein the at least one processor estimates the signal and on-time interference statistics with a first filter having a first bandwidth, and estimates the noise and multipath interference statistics with a second filter having a second bandwidth.

30. The apparatus of claim 29, wherein the at least one processor selects the first and second bandwidths based on channel conditions.
31. A method comprising:
obtaining samples from multiple receive antennas for a multiple-hiput multiple-output (MTMO) transmission sent from multiple transmit antennas;
estimating signal and on-time interference statistics based on the samples; estimating noise and multipath interference statistics based on the samples; and
estimating signal, noise and interference statistics based on the estimated signal and on-time interference statistics and the estimated noise and multipath interference statistics.
32. The method of claim 31, further comprising:
deriving first channel estimates based on the samples; and
filtering the first channel estimates to obtain second channel estimates.
33. The method of claim 32, wherein the estimating the noise and multipath
interference statistics comprises
differentiating the first channel estimates to obtain differentiated results, and estimating the noise and multipath interference statistics based on the differentiated results.
34. The method of claim 32, wherein the estimating the signal and on-time
interference statistics comprises
estimating the signal and on-time interference statistics based on the second channel estimates.
35. An apparatus comprising:
means for obtaining samples from multiple receive antennas for a multiple-input multiple-output (MIMO) transmission sent from multiple transmit antennas;
means for estimating signal and on-time interference statistics based on the samples;
means for estimating noise and multipath interference statistics based on the samples; and
means for estimating signal, noise and interference statistics based on the estimated signal and on-time interference statistics and the estimated noise and multipath interference statistics.
36. The apparatus of claim 35, further comprising:
means for deriving first channel estimates based on the samples; and means for filtering the first channel estimates to obtain second channel estimates.
37. The apparatus of claim 36, wherein the means for estimating the noise
and multipafh interference statistics comprises
means for differentiating the first channel estimates to obtain differentiated results, and
means for estimating the noise and multipath interference statistics based on the differentiated results.
38. The apparatus of claim'36, wherein the means for estimating the signal
and on-time interference statistics comprises
means for estimating the signal and on-time interference statistics based on the second channel estimates.
39. An apparatus comprising:
at least one processor to obtain samples from multiple receive antennas for a multiple-input multiple-output (MEMO) transmission sent from multiple transmit antennas, to determine channel conditions based on the samples, to select one of multiple channel and noise estimation schemes based on the channel conditions, and to perform channel and noise estimation based on the selected channel and noise estimation scheme; and
a memory coupled to the at least one processor.
40. The apparatus of claim 39, wherein for the selected channel and noise estimation scheme the at least one processor derives channel estimates based on the samples, estimates signal and on-time interference statistics based on the samples, estimates noise and multipath interference statistics based on the samples, and estimates signal, noise and interference statistics based on the estimated signal and on-time interference statistics and the estimated noise and multipath mterference statistics.
41. The apparatus of claim 40, wherein the selected channel and noise estimation scheme is used if the channel conditions indicate a small delay spread, a single path environment, low velocity, or a combination thereof.
42. The apparatus of claim 39, wherein for the selected channel and noise estimation scheme the at least one processor derives channel estimates based on the samples, estimates total received energy based on the samples, estimates signal and interference energy based on the channel estimates, and estimates noise based on the estimated total received energy and the estimated signal and interference energy.
43. The apparatus of claim 42, wherein the selected channel and noise estimation scheme is used if the channel conditions indicate a large delay spread, a multipath environment, high velocity, or a combination thereof.
44. The apparatus of claim 39, wherein for the selected channel and noise estimation scheme the at least one processor derives channel estimates based on the samples and estimates signal, noise and interference statistics based on the samples.
45. The apparatus of claim 39, wherein the at least one processor determines delay spread in a communication channel between the multiple transmit antennas and the multiple receive antennas, and selects the channel and noise estimation scheme based on the delay spread.

46. The apparatus of claim 39, wherein the at least one processor determines energies of signal paths in a communication channel between the multiple transmit antennas and the multiple receive antennas, and selects the channel and noise estimation scheme based on the energies of the signal paths.
47. The apparatus of claim 39, wherein the at least one processor selects the channel and noise estimation scheme based on estimated velocity.
48. A method comprising:
obtaining samples from multiple receive antennas for a multiple-input multiple-output (MIMO) transmission sent from multiple transmit antennas;
determining channel conditions based on the samples;
selecting one of multiple channel and noise estimation schemes based on the channel conditions; and
performing channel and noise estimation based on the selected channel and noise estimation scheme.
49. The method of claim 48, wherein the performing channel and noise
estimation comprises
deriving channel estimates based on the samples,
estimating signal and on-time interference statistics based on the samples, estimating noise and multipath interference statistics based on the samples, and estimating signal, noise and interference statistics based on the estimated signal
and on-time interference statistics and the estimated noise and multipath interference
statistics.
50. The method of claim 48, wherein the performing channel and noise
estimation comprises
deriving channel estimates based on the samples, estimating total received energy based on the samples,
estimating signal and interference energy based on the channel estimates, and estimating noise based on the estimated total received energy and the estimated signal and interference energy.
51. The method of claim 48, wherein the selecting one of the multiple
channel and noise estimation schemes comprises
selecting one of the multiple channel and noise estimation schemes based on delay spread, path conditions, velocity, or a combination thereof.
52. An apparatus comprising:
means for obtaining samples from multiple recerve antennas for a multiple-input multiple-output (MEMO) transmission sent from multiple transmit antennas;
means for deterniining channel conditions based on the samples;
means for selecting one of multiple channel and noise estimation schemes based on the channel conditions; and
means for performing channel and noise estimation based on the selected channel and noise estimation scheme.
53. The apparatus of claim 52, wherein the means for performing channel
and noise estimation comprises
means for deriving channel estimates based on the samples,
means for estimating signal and on-time interference statistics based on the samples,
means for estimating noise and multipath interference statistics based on the samples, and
means for estimating signal, noise and interference statistics based on the estimated signal and on-time interference statistics and the estimated noise and multipath interference statistics.
54. The apparatus of claim 52, wherein the means for performing channel
and noise estimation comprises
means for deriving channel estimates based on the samples,
means for estimating total received energy based on the samples,
means for estimating signal and interference energy based on the channel
estimates, and
means for estimating noise based on the estimated total received energy and the
estimated signal and interference energy.
55. The apparatus of claim 52, wherein the means for selecting one of the
multiple channel and noise estimation schemes comprises
means for selecting one of the multiple channel and noise estimation schemes based on delay spread, path conditions, velocity, or a combination thereof.

Documents:

http://ipindiaonline.gov.in/patentsearch/GrantedSearch/viewdoc.aspx?id=LPHkN9ILPwCvRJUmAIifMA==&loc=+mN2fYxnTC4l0fUd8W4CAA==

« Previous Patent

Next Patent »

Patent Number

268261

Indian Patent Application Number

3485/DELNP/2008

PG Journal Number

35/2015

Publication Date

28-Aug-2015

Grant Date

24-Aug-2015

Date of Filing

28-Apr-2008

Name of Patentee

QUALCOMM INCORPORATED

Applicant Address

5775 MOREHOUSE DRIVE, SAN DIEGO, CALIFORNIA 92121-1714, U.S.A.

Inventors:

#	Inventor's Name	Inventor's Address
1	BYOUNG-HOON KIM	8775 COSTA VERDE BOULEVARD, #912, SAN DIEGO, CA 92122, U.S.A.

PCT International Classification Number

H04L 1/06

PCT International Application Number

PCT/US2006/060372

PCT International Filing date

2006-10-30

PCT Conventions:

#	PCT Application Number	Date of Convention	Priority Country
1	60/731,423	2005-10-28	U.S.A.
2	11/553,296	2005-10-26	U.S.A.