TWI345895B - Multi-stage receiver for wireless communication - Google Patents

Description

1345895 IX. Description of the Invention: [Technical Field of the Invention] The present disclosure relates to communication 'and more specifically to techniques for receiving a multiple input multiple output (MIMO) transmission. [Prior Art]

The transmission is transmitted from multiple (Μ) transmit antennas to multiple (ν) receive antennas. For example, the t: 'transmitter can simultaneously transmit the job stream from one transmit antenna. The A f data stream is degraded by the line environment & true and further by noise and interference. The receiver receives the transmitted data grab via the N receive antennas. The signal received from each receive antenna contains a scaled and delayed version of the transmitted data stream. The transmitted data stream is thus dispersed among the N received signals from the N receive antennas. The receiver then processes the N received signals by a space-time equalizer to recover the transmitted data stream. The receiver can dynamically derive the coefficients of the space-time equalizer to account for changes in signal characteristics. These signal characteristics can be related to channel and interference statistics, (4) space-time processing of data streams, and the like. The calculation of the equalizer coefficients is computationally intensive. Updating these equalizer coefficients to match the fastest changes in signal characteristics can result in very complex receivers. Updating these equalizer coefficients at a slower rate can result in performance degradation. Therefore, there is a need in the art to efficiently receive the technology of MlM transmission. SUMMARY OF THE INVENTION Techniques for efficiently receiving M1M0 transmissions are described herein. The receiver receives the plurality of signals transmitted by the transmitter and processes the plurality of received signals to obtain the received data. The receiver can process the received data in multiple stages. Secondly, the front-end chopper performs a money/equalization (eg, 'compensation, suppression, or mitigation') of the non-punctual signal component of the received material. The second stage is composed of one or more combiners. Component. Punctual and level-of-four (4) wave data can be distinguished based on the time of the group and non-punctual signal components. At the receiver, the on-time signal component can include other symbols that are simultaneously transmitted with the desired symbol. Here, the time component may include a component that is not a component of the on-time signal = such as a signal component that is tracked to a complex symbol that is transmitted before and after the desired symbol. ', the front-end filter does not see the cup 南 $ south gauge emission 15 The data is specific to the processing, and the combiner matrix can be a function of the data specific processing at the transmitter. The processing can correspond to, for example, a channelization code, a transmission matrix, a gain, etc. for transmitting data.兮, less ° Using the code division multiplexing (CDM) ΜΙΜΟ transmission, you can get a single-pin 4 mountain 4 early 鳊 filter and use it for all channelization 1 and can get different combiner matrix For the per-channelization code, the receiver processing is divided into multi-level servants to simplify the "end filter and combiner matrix" while achieving good performance. The end filter and combiner matrix are updated separately at the same or different rates. According to one aspect, a device is described. The processor performs filtering on the received data: the device and the body, and processes the non-punctual signal components of the plurality of received data and obtains the filtered data. The processor also processes the filtered data to signal a plurality of transmitted signals # and align the time. According to another aspect, a device comprising at least one processor and a memory I16763.doc 1345895 is described. The processor derives a front end filter for processing non-punctual signal components of the plurality of received signals and also derives a plurality of combiner matrices for puncturing the plurality of transmit signals transmitted by the plurality of channelized codes Signal component. The processor filters the received samples by a front end filter and obtains filtered symbols of the plurality of channelized codes. The processor further processes the filtered symbols of each channelization code by a combiner matrix for the channelization code to obtain an output symbol of the channelization code. Various aspects and features of the present disclosure are described in further detail below. [Embodiment] The receiver processing techniques described herein can be used in various communication systems, such as a code division multi-directional proximity (CDMA) system, a time division multi-directional proximity (TDMA) system, a frequency division multi-directional proximity (FDMA) system, and orthogonality. FDMA (OFDMA) system, single carrier FDMA (SC-FDMA) system, and the like. CDMA systems utilize code division multiplexing (CDM) and transmit modulation symbols in parallel using different channelization codes. CDMA systems may utilize radio technologies such as Wide Frequency CDMA (W-CDMA), cdma2000, and the like. Cdma2000 covers IS-2000, IS-856 and IS-95 standards. TDMA systems may utilize radio technologies such as the Global System for Mobile Communications (GSM). W-CDMA and GSM are described in documents from an organization called "3rd Generation Partnership Project" (3GPP). Cdma2000 is described in a document from an organization called "3rd Generation Partnership Project 2" (3GPP2). 3GPP and 3GPP2 documents are publicly available. An OFDMA system utilizes orthogonal frequency division multiplexing (OFDM) and transmits modulation symbols in the frequency domain on orthogonal subcarriers. The SC-FDMA system utilizes single carrier frequency division multiplexing (SC-FDM) and transmits modulation symbols in the time domain on the orthogonal subcarriers. 116763.doc 1345895 The techniques described herein are also applicable to MIM〇 transmissions on the downlink as well as on the uplink. The downlink (or forward link) refers to the communication link from the base station to the wireless device, and the uplink (or reverse link) refers to the wireless device.

Figure 1 shows a block diagram of a transmitter 11 and a receiver 15 for transmission. For downlink transmission, the transmitter 11 is part of the base station and the receiver 15G is part of the wireless device. For uplink transmission, the transmitter 11G is part of a wireless device and the receiver (9) is part of a base station. A base station is typically a fixed station that communicates with wireless devices and can also be referred to as a node, an access point, and the like. A wireless device can be stationary or mobile and can also be referred to as a user equipment (UE), mobile station, terminal, station, subscriber unit, and the like. The wireless device can be a cellular phone, a personal digital assistant (pDA), a wireless data modem, a laptop, a palmtop device, and the like.

Prepare the communication key to the base station. For clarity, the techniques for MIM(R) transmission in a CDMA system are described below, which may be ^w_cdma, cdma2000 or some other CdmA radio technology. At transmitter 110, a transmit data processor (τ χ data processor) 112 receives the received data from the traffic data and signals, processes (e. g., encodes, interleaves, and symbol maps) the received data and provides the data symbols. Processor 112 also generates and multiplexes pilot symbols having data symbols. As used herein, a data symbol is a symbol used for traffic data or signal transmission, a pilot symbol is a symbol for a pilot, and the symbol is typically a composite value. The data symbols and pilot symbols can be modulation symbols from a modulation mechanism such as PSK or QAM. The pilot is a known prior to the transmitter and receiver. The processor 114 performs spatial or spatial-time processing on the data and pilot symbols as described below, H6763.doc 1345895 and provides the output symbols to multiple ( M) CDMA modulator 116 & to U6m. Each CDMA modulator 116 processes its output symbols as described below and provides the output chips to an associated transmitter unit (TMTR) 丨i 8. Each transmitter unit 118 processes (e. g., converts to analog, amplifies, filters, and upconverts) its output chips and produces a modulated signal. M modulated signals from M transmitter units 118a through 118m are respectively transmitted from M antennas at "to 12 jaws. At receiver 150, multiple (N) antennas 152 & to 15211 are in a wireless environment. The transmitted signals are received via various propagation paths, and the N received signals are respectively provided to N receiver units (11 (; 11 11) 1543 to 15 牝. Each receiver unit 154 processes (eg, filters, amplifies, Down-converting and digitizing) the received signal and providing the received samples to the channel processor 156 and the equalizer/CDMA demodulator 160. As described below, the processor 156 is derived for the front-end filter/ The coefficients of the equalizer and the coefficients used for the one or more combiner matrices. The single το 160 performs equalization by the front-end filter and CDMA demodulation, and provides filtered symbols. Receive (RX) MIM 〇 processor [Μ Combining the filtered symbols on the spatial size and providing the detected symbols, the detected symbols are the estimates of the transmitted data symbols. The Rx data processor 172 processes (eg, 'symbol demapping, deinterleaving, and decoding) the extracted symbols and Provided decoded In general, at the transmitter 110, the processing by the equalizer/CDMA demodulator 160, the RX MIM〇 processor 17〇& RX data processor 172 is supplemented by the CDMA demodulator ιΐ6, τχ The processor II4 and the TX data processor U2 perform processing. The controller/processors 13A and 180 respectively instruct the operation of various processing units at the transmitter 11 and the receiver "Ο 116763.doc • 10· 1345895. 132 and 182 store data and code for the transmitter 110 and the receiver 150, respectively.

Figure 2 illustrates a MIMO-CDM transmission. For CDM, up to C symbols can be simultaneously transmitted via a transmit antenna via C channelization codes, typically C21. Such channelization codes may be orthogonal variable spreading factor (OVSF) codes in W-CDMA, Walsh codes in cdma2000, other orthogonal codes or quasi-orthogonal codes, pseudo-random codes, etc. . Each channelization code is a chip of a particular sequence. The number of chips in the sequence is the length of the channelization code or the spreading factor. In general, one or more channelization codes of either set can be used for each transmit antenna, and the channelization codes can have the same or different spread spectrum factors. For the sake of simplicity, the following description assumes that the channelization codes have the same spreading factor. The C channelization codes of the same set can be reused for each of the two transmit antennas. For ΜΙΜΟ, up to 符号 symbols can be sent simultaneously via one transmit antenna. For MIMO-CDM, up to 符号 symbols can be transmitted simultaneously via C transmit codes via one transmit antenna. The ΜΙΜΟ processing can be performed separately for each of the C channelization codes. Perform ΜΙΜΟ processing on each of the two transmit antennas for each channelization code. CDM processing can be performed separately for each of the launch antennas. For each transmit antenna, CDM processing is performed on all C channelized codes. Figure 3 shows a block diagram of a CDMA modulator 116 for a transmit antenna m. A CDMA modulator 116 can be used for each of the CDMA modulators 116a through 116m of FIG. The CDMA modulator 116 includes a data processor 310 for each channelization code for traffic data and/or signal transmission and a pilot processor 320 for pilots. 116763.doc -11 - 1345895 In the data processor 310, the spreader 312 spreads the output symbol for the data by using the channelization code c of the chip sequence training ((4), where 4 symbol index ' and k is the code The slice index multiplexer 314 scales the round-trip of the spreader 312 with the benefit gmc and provides a data chip for the channelization code c. Within the pilot processor 320, the spreader 322 is used by the pilot The channelized code P of the frequency is used for spreading the output symbol 'force' of the pilot. The multiplier 324 boosts the output of the spreader 322 by scaling, and provides pilot chips. The gain %/ and the amount of transmit power used for the channelized code and the pilot are determined separately. Summer 330 sums the data for all channelization codes with the pilot chips. The scrambler 332 combines the output of the summer 33 with the scrambling sequence P for the transmitter 11 (phantom multiplication and provides the output chip for the transmit antenna ym{k) ° in general, c Any number and any number of channelization codes can be used for each of the transmit antennas. In one design, the same channelization code is used for pilots for all M transmit antennas. In another design, for each transmit antenna, M channelization codes are used for the pilot, and the remaining c·one channelization codes can be reused for each of the plurality of transmit antennas. As shown in Figure 3, the same scrambling sequence can be used for all of the transmit antennas. Alternatively, _ different scrambling sequences can be used for each transmit antenna. Spreading and scrambling can also be performed in other ways. The channel is formed by the propagation environment between the μ transmit antennas at transmitter U0 and the one of the receive antennas at receiver 150. For each-channelized code, one data symbol can be sent in parallel from one transmit antenna, where hL^nin{M, Ν} » receiver 150 can have different values for L (and possibly different 116763.doc -12- The transmit matrix/vector) analyzes the performance of the MIM0 channel (eg, transports the magic and selects the L value for the best performance (and the transmit matrix/vector channelization code c is executed to the transmitter 110 for each symbol period of $ for each The following transmitter spatial processing: Equation (1) Lx = of the data symbol, c = 7,..., C > where: L(4) = [U5)62 c ("(4)f is the quantity, and the °° is used for channelization The LxM emission matrix of code c, gross (4)H(4) is the ship vector of the output symbol and, T" main-M_ table does not move the item. (10) Each element can correspond to - different data streams. The data stream can have S: "Yes in this In this case, matrix I can have different row standards for different data streams. Equation (1) can be spatially coded by L. It can also perform spatial time exution such as space-time transmit diversity (STTD), but it is not shown in equation (1). ). Different emissions (4) can be used in ^ (four) fine mode, such as closed loop emission points Set (CLTD), Daily Line Rate Control (PARC), Code Reuse Bell Lab Space/Time (CRBLAST), Dual Launch Adaptive Array (D_txaa), etc. Table 1 lists some MIM0 modes, and for each The mode gives the L, M, 屯 and data symbol sources. In the table μ u is selected from the set " β'π ] Γ > [1 β^π/4γ , μ eJ^/4, T , η .π / 2x2 matrix jit / Λ _-y'3n / 4, 73π / 4 J L1 6 J t1 e7 ] 7 2 parent 1 vector. L-iJtaa can be selected from the set of ~ diagonal is one and in other Unit matrix of zero. 116763.doc •13- 1345895 Table 1 ΜΙΜΟ mode LM Be data symbol source CLTD 1 2 from a single coded frame. PARC L=M > 2 Bc=i from L different coded frames CRBLAST L = M > 2 Sc = I from a single coded frame. D-TXAA L = M = 2 from up to L coded frames. Transmitter 110 can transmit for each symbol period S The antenna w performs the following CDMA processing: ^ (^) = Σ Sm,c vc (k mod C) · dmfi (k div C) I · P(k) C=, ”, W = 1, _', M, equation (2) where the symbol period ^ corresponds to that given by S = kdivC = Lk/C" Sheet moxibustion period for each channel code 'Unused gain heart. It can be set equal to the stone disaster. The CDMA processing with different spreading frequency can be expressed as: ym^ 'Vc^k modCc)'dmAk div Cc)\p(/t) ^ , where Cc is the channelization code e Frequency factor, and NPC is the number of channelization codes used for the transmit antenna. For the sake of simplicity, the following description assumes a channelized 赞 射 antenna that exhibits a frequency factor C for each transmission. In the equation, the channelization code e with the spreading factor c is used for transmission and scaling to obtain data chips. The gain is achieved by multiplying the gain by the ratio of the next output symbol ^, the symbol of multiple c-coded chips, illusion 116763.doc -14- 1345895. The data and pilot chips for all c channelization codes are summed and further scrambled with a scrambling sequence to obtain the output chip %(A:) for the transmit antenna m. The same CDMA processing is performed on each of the river transmitting antennas. At each chip week, the sample received at the receiver 150 in the moxibustion can express i.(k))=E2L(k) + n(k), Equation (3) where: 31 (magic output chip Txl vector, where τ is described below, Η is the RXT channel response matrix, where the ruler is described below, 1 (none) is the Rx 1 vector of the received sample, and 1L (moxibustion) is the Rx 1 noise direction. 150 may multiply the signal received from each of the receiving antennas, where K is the oversampling rate and is generally. In each moiré cycle, the receiver 150 may obtain Ε from each of the receivers 154. κ samples, and by the N receivers 154 & 154n stacked Ν·Ε.Κ samples to form hair (moxibustion). Ε is the number of chips in the receiver 15 之前 front end equalizer ^, α, EU and can be chosen based on the trade-off between receiver complexity and performance. Nail illusion includes R received samples from \ receiving antennas with a chip period, where R = nek. Used for time domain channel impulse responses for all transmit and receive antenna pairs, as shown in Figure 1, at each transmit antenna and each connection There is a propagation channel between the antennas, or there are a total of MN propagation channels between the M transmitting antennas and the N receiving antennas. Each of the propagation channels has a wireless 116763.doc •15· yj ^ j environment determined > 4dt + Line heart pulse response. Each transmit antenna receives the day... The response of the early input multiple output (SIMO) channel can be given by the μ array Sm. Η Λ Λ Η Η ^ — ^ The number of columns is 1 (岣中The number of inputs is determined. The number of bits is determined by the time interval between the equalizer length Ε and the impulse response between the transmitting antenna and the receiving days. L can be given as follows:

Ε + Equation (4) /»», "The interval between the impulse response between the transmitting antenna and the receiving antenna" in terms of the number of snippets and gentlemen + γ 1 . As follows, 'matrix _ submatrix &, where =1,...,m: equation (5) Η has the size of RxT, where τ=τι+τ„ +Τ Μ for one transmit antenna, vector (10)) contains Difficult vector l(4), where w 1,···,Μ. Each sub-vector ^ (free) includes the output chips concentrated in the chip period, from a transmitting antenna. Vector (phantom and sub-vector) l(4) can be expressed as: 'LW ^(^-LTm/2-lJ)- yW = to) and = ym(k) MW_ .ym(k + [Tm/2-l\) Equation (3) can also be expressed For: Equation (6) χ^)=Σηη, ι„,^) +sw 〇 Equation (7) m»l 116763.doc -16 - For the model shown in equation (7), in each chip shot The Tm output chips are sent from each of the transmitting antennas to the N receiving antennas via a WMO channel and a response from the WMO channel. Jade (the sample received by the Magic Center includes contributions from all of the transmitting antennas), and this can be relatively large. As an example, in the case of Μ=2, Ν=2, κ=2, Ε=20, Τ=48, and R=80, L (the illusion can be judged as “the vector can be 8 〇 to 8 matrix” and 2_(The magic can be 8〇χι vector. The noise can be assumed to be a fixed compound random vector, where equation (8) E{n(1) is 0, and equation (9) where five { } is the expected operation and 9 is all zeros. Vector, the m*RxR noise covariance matrix, and the "heart" indicates the pairwise shift term. Equations (8) and (9) indicate that the noise has a zero mean and a covariance matrix for the region. The receiver 15 0 can be borrowed as follows Recovering Ks for each channelized code c by a set of [chopper-paired samples for the channelized code] and then de-spreading and descrambling the filtered samples In the data symbol: λ 1 (i+l)C—1 ^C(J) —c iW'[vc(^modC)-^(it)]' Equation (10) Equation (11) = See [None X(A) + B(10)· [ve(A: mod C} · p(A)]* =Wf [H0c(5) + nc(i)] =Wf xc(s) fi Γί+υ〇-ι c W = into β m〇d cj · 〆切·116763.doc -17- 1345895

i W · [vc (k mod C)p{k)}' = H 0C (j) + nc (s) Equation (12), Equation (13) and 1^11 for channelization code <: An overall filter, estimated for kcb), and "*" indicates composite pairing. Color (0 is the Txl vector used to decode the spread spectrum symbol of the channelization code and is obtained based on the transmitted chip of #. acG) is the RX1 miscellaneous after the descrambling and despreading of the channelization code c Signal vector. L(4) is saved (the statistic of magic, which is independent of the channelization code C. ^(J) is the RX1 vector for the despread symbol of the channelization code C and is obtained based on the received samples. ^included for channelization The L filters of the S group of the code c. The equation (1 〇) indicates that the symbol can be equivalently executed by the processing instead of the received sample in the 1 (moxibustion). The filter I can be Weiner A filter, which can be derived as follows:

Wc = E{ zc(s) £ (ί)}-1 ·Ε{χ£ (s) bf (5)} =[Η £ {Θε(s) 0f (s)} KH + ]-' · [H £ {0C(s) bf (s)}]

= [HrcHw+Rnn]-1HL Equation (14) Equation (15) = Rc1Hrc=R;1 Φ, where , Equation (16) Equation (17)

E{0c(s)b?(s)} » X

Rc = HrcHn+R 116763.doc -18- 1345895 £c is the covariance matrix of the color (1). l is the TxL correlation matrix of colors (4) and L〇) and can indicate the characteristics of the transmitted signal. & is the RxR correlation matrix for wireless channels, signaling and noise systems. The matrices X, Xc and Kc are determined by the transmission matrix I and the gain depending on the channelization code C. The matrix £c can be expressed as: Equation (18) where I] is He and does not depend on the channelization code c, one part of β can be shown, and only one element of l is non-zero. The exponent of the non-zero element makes the RxL matrix I can be expressed as:

Hyc =〇c = HGCBc > Equation (19) where: H is the RXM matrix containing the \4" punctuality" rows, and is the ΜX Μ diagonal matrix containing V^.gm,c along The diagonal w=l,...,M and is zero elsewhere. As shown in equation (5), g_ contains Τ = Τι + Τ 2+_·· + Τ μ rows of 8 beans containing Η for the transmitting antenna, or one row for each transmitting antenna. The row is multiplied by the element of the phantom that has its chip time aligned with the solution spreader operation in equation (1〇). These lines can be identified as described below. The matrix in equation (17) can also be expressed as:

Be =nrcRH+Rm

= +Rm+nlclHcnH = Ε+Φε of Equation (20) 116763.doc -19· 1345895 where R = HEHH+Enn. The matrix I in equation (14) can then be expressed as equation (21)

Wc =Κ; 'ΗΟ,Β0 = [! + 〇, Φ^Γ'ΗΟ,Β, =R"' HGc Bc (I + Bf Gc Hw R1 HGc BJ1 = FAC where: E = I -1H, and the equation ( 22) Equation (23)

Equation (24) In equation (23), ^ is a relatively large RXM matrix that does not depend on the channelization code. In equation (5), ^ is a smaller MxL matrix containing a matrix of all the dependent codes in the equation. Equations (10) through (4) indicate that the processing at receiver 15() can be performed in two stages. The first stage is filtered by the sample received by the front end filter without depending on the channelization code, and the sample of the (four) wave is despread and descrambled to obtain the sign of the transit wave. Single-front-end filters and filters can be used for all channels. The second stage separately updates the front-end filter at the same rate or at different rates for each channel-based code and the combination of the symbol and the combiner to obtain the detected symbols for the channel. Combiner matrix. Multi-stage receiver processing can be performed in various ways. In the following description, the pilot symbols are assumed to be "transmitted by H" and transmitted for the same channelization code p of m transmit antennas. The pilot symbols are also assumed. For the parent, make P(8) both (8) b!, which is the Mxl vector of the pilot symbol transmitted from one transmit antenna in the period 116763.doc • 20-number period ί. In a receiver design, the front end is obtained. Filter I and use it for the first stage (eg, 'for block 160 in Figure 1), and the combiner matrix & is calculated for each channelization code and for the second level (eg, For block 170 in Figure 1) For symbol level training, the filter can be derived based on the pilot symbols using the least squares criterion as follows: ^p = 1Sin Σ ιι^χρω-^ωιι2 · Equation (25) : Heart (4) is the Rxl vector of the despread pilot symbol ^ is the RxM candidate filter matrix, the bundle p is the RXM filter matrix derived based on the pilot symbols, and P is the number of pilot symbols used to derive. Although pilot channelization codes are used instead of channelization codes c, but the despread pilot symbols in & (1) can be obtained as shown in equation (13). The least squares criterion selects a candidate filter matrix from all candidate filter matrices, minimizing the number of right-hand sides. The selected filter matrix is provided as the least squares filter matrix %. In the limit case, when P - oo in equation (25), % can be expressed as: Equation (26) yVp = [E{xp(s)zHp (s)}Vl E{xp(s)bHp(s)} 〇 For symbol level training, it can be obtained as follows: The despread pilot symbols in ' as shown in equation (13) can be received from Obtained from the sample. It can be calculated and averaged RXR outer product for a sufficient number of pilot symbols 116763.doc -21- (8) know (s). It can also be calculated and averaged RXM outer product (8) are all (8). ' 'The calculation of the outer product can also be based on the recursive least squares () block least squares or this technique Some other techniques known in the art result in I. For chip level training, the filter can be derived based on the received samples using the least squares criterion as follows: (Jo+P)Ci —p plexus c 11 ))-fep(7)·νρ(Α:ιη〇(ΙΟ)·/7(Λ)||2 . , Equation (27) 八中b〇) (The magic is by spreading and scrambling the pilot symbols) Know the Μχ 1 vector of the pilot chip. In the limit case, in equation (27), % can be expressed as: I, (4) from i), gas equation (10) for chip level training, I is obtained as follows. Based on the received, the RxR outer product is expressed as "(1) and averaged for a sufficient number of pilot symbols. It is also possible to calculate and average the RxM outer product (1)~α). The enthalpy can then be calculated based on the two averaged outer products. From equation (22), 1 can be expressed as: wP = rHGp(i + GPrrHG r* - p ) = ια Ρ α + - ρ Ρ τι Equation (29) Equation (30) where: L = irrsSp.

The MxM matrix l can be estimated as D: 1 J〇 + P - 1 £ρ = 7 · Σ Α SS 〇 equation (31) 116763.doc -22- 1345895 In the extreme case, P - 〇〇, the estimated change of b More precise. Similar to I, it is also possible to estimate 匕 based on pilot chips instead of pilot symbols. The front-end chopper can be derived based on equation (29) as follows: Equation (32) The combiner matrix core can be derived based on equations (24) and (30) as follows: Equation (33)

The filter & can be based on ^ and b as follows:

Wc =EGCBC(I + Bf · =3⁄4(卜玉(卜 equation (34)

As shown in equation (32), the front end filter E can be derived based on ^ and b and can be estimated based on pilot symbols or chips. As shown in equation (33), the combiner matrix for each channelization code can be based on £p, the gain matrices b and ^ for pilot and data, and the transmit matrix for channelization code c. I got it. t is also known as the traffic to pilot ratio and can be known (e.g., 'transmitted via a signal) or estimated by the receiver. Estimating the ratio of traffic to pilot is usually sufficient and there is no need to separately estimate & and 1. Receiver 150 can recover the data symbols in LQ) as follows: tCO = 乂 [eg (+f V _ 憎 m〇d 〇 · _]·] W i=iC ). Equation (35) The receiver 150 can perform the processing of Equation (35) in two stages. In the first stage, I16763.doc -23- 1345895. In the first part, the receiver 150 can perform the mountain filtering in each chip period as follows: YW=FHx(k). Equation (36) where χ(Ατ) For the Mx 1 vector of the wave sample, in the second part of the first stage, the receiver 150 can despread and descramble the filtered samples for the following _ ^ channelization code c: Λ | (ί+丨)C-1 -c^ = _/n' Σ v(A:)-[vc(^modC)-/7(^)]* • λ *=5C , equation (37) where 士士(7) is filtered The Mxl vector of the symbol, which is an estimate of ^(ί).

In the second stage, the receiver 150 can be combined for each channelization code as follows: T Equation (38) The receiver 1 50 can also recover the data symbols in the following:

H〇(s) = Af fw .(+^(k). [Vc{k m〇d C). p{k)]. *=fC . Equation (39) can also be performed in two stages (39). In a first portion of the first stage, receiver 150 may despread and descramble the received samples for each channelization code c to obtain despread symbols for the channelization code. In the first, second portion, the receiver 150 may perform filtering on the despread symbols for each channelization code c by the same front end filter F to obtain filtered symbols for the channel, code. In the second stage, the receiver (10) can filter the symbols for each channelization code e with the combiner matrix 4 for the channelization code. Add a new _. 116763.doc -24- 1345895 Private is obtained based on pilot symbols or chips and may not be aggregated to the optimal setting' (for example) because a limited number of pilot symbols (p) are used for training. In such cases, the use of the U-based and how can lead to performance degradation. In another receiver design, \y Tian Tiantian Silver & is used as the first stage front end filter. The combiner matrix is also calculated for the helmet, s .. ., rff I Dan for the parent one channelization code C and for the second level 0

The chopped symbol obtained by ^^ can be expressed as: ?c(-s) = yiHP xc{s) = W^HGC Bc bc(i) + nc(5) = Ae bc(j) + ne(i Where: ac = < h^Sc, and the data symbol in the filtered symbol kc(*s) for the channelization code c can be obtained as: Μχ 1 vector of equation (40) Equation (41).

Fec(i) = Df zc(j), where equation (42) is the MχL combiner matrix for the channelization code c. Can be derived based on the Minimum Mean Square Error (MMSE) criterion as follows: ° The moment SC = (AC + U4C. The noise covariance matrix is also) can be estimated by adding Equation (43) as follows: 116763.doc •25· 1345895 j J〇+Pl -mC = P ' s^cl~p{S)'hp(s)] [ZM- A,(,)b/.)]" _ 1 ～ Equation (44) Among them, drink HQ, the second equation in equation (44) is derived from the unbiased estimation of the heart is only the fact.

The MxL matrix heart can be expressed as: Equation (45)

As shown in equation (45), it can be based on (4) self-pilot symbols or chip estimates and can be applied to the matrices of all channelized codes and (8) traffic-to-pilot ratios specific to the channelized code c g &amp And the emission matrix i and the matrix 4C for each channel. As shown in equation (43), it can be calculated based on (4) a noise covariance matrix u (b) applicable to all channelization codes, a matrix & for channelization code c, and a calculation for each channelization code Combiner matrix 込.

The combiner matrix can also be estimated for each channelization code e: C = 1 ' and Equation (46) Equation (47) where EzZ is the covariance matrix for L(J). In equation (46), the product ιωβω for each channelization code can be calculated, then averaged for all channelization codes with the same emission matrix, and further averaged for a sufficient number of symbol periods. Knowing the covariance matrix » can then be calculated based on the reciprocal of the heart and also. 116763.doc -26 · Receiver 150 can recover the data symbols as follows: fic(^) = Dc ^~ι=· ^ W^x(A:) [vc(A:modC) p(jfc)]* iC ). Equation (48) In equation (48), the receiver 15〇 can chop the received samples by the front-end filter, and then despread and descramble the chopped samples for each channelization code c. The frequency, and then the filtered symbols and combiner matrices for each channelization code are also combined. The receiver 150 can also recover the data symbols as follows: he(s) = Dc WP -= · £ x(A:). [Vc{k mod C) · ^(yfc)]* *=iC . Equation (49) In equation (49), the receiver 15A can despread and descramble the received samples for each channelization code c, and then use the front-end filter % pair for each channelization. The decoded spread symbols of the code are filtered, and then the filtered symbols for each channelized code are combined with the combiner matrix. Receiver processing can also be performed in multiple stages in other ways. Described below - specific receiver design. 4 shows a block diagram of one of the various blocks within the receiver 150 of FIG. In this design, front-end filtering is performed prior to CDMA demodulation. Block 160 includes a front end filter/equalizer 410 and a CDMA demodulator 420. Demodulator 420 includes C descrambler/despreaders 422a through 422C for interfacing C channelized codes for traffic data. Block 170 includes C combiners 432a through 432C for up to C channelization codes for traffic data. Within channel processor 156, timing estimator 442 determines 116763.doc -27 - 1345895 of the received signal, timing unit 442 can estimate the channel impulse response and/or power. delay profile for the different antennas, and can determine the channel impulse response and / or the center of gravity of the power delay profile. Unit 442 can then determine the timing of the received signal based on the center of gravity. The coefficient calculation unit 444 derives the coefficient of the chopper % based on the received samples, for example, as shown in the equation (28). Unit 444 can also be derived based on a similarity, block least squares, or some other technique. Timing from Unit 2 • Information can be used for training, for example, to align locally generated pilot chips with received samples. Unit 444 provides ^ to front end filter 41A. Filter 410 performs front-end chopping/equalization by the sample training received by the private pair and provides filtered samples. Within the demodulator 42A, each unit despreads and descrambles the filtered samples for a different channelization code and provides filtered symbols for the channelized mom. The unit 446 despreads and descrambles the filtered samples for the pilot channelization code p. Units 422 and 446 perform despreading and descrambling based on the timing provided by unit 442. Channel estimator 448 estimates the MxM matrix based on the filtered pilot symbols from unit 446. 2. Unit 45 计算 computes the outer product 7(7)^(7) for each channelization code, averages the outer product for all channelization codes and symbol periods, and provides a correlation matrix of 2, for example, as in equation (46). Unit 452 derives the coefficients of the combiner matrix for each channelization code c based on the matrix from unit 448 and the correlation matrix from the unit 兀 450 and the code specific matrix as follows: Equation (50) 〇. = This =: ^ 昼仏 Bc. 116763.doc • 28· 1345895 Within the RX® processor η0, each combiner 452 combines the filtered symbols for a different channelized code c based on the combiner matrix and provides a pass for the channelized code The symbol of the detection. Receiver 150 can estimate the quality of the received signal, which can be quantified by signal interference noise ratio (SINR) or the like. The detected symbols from equation (42) can be linked as:

^c(5) = DC Ac + nc(j) = hc l?c(-y) + wc(i) Equation (51) where Lf and the corpse are 仏 (1). The covariance of lCs) can be given as = ugly _) s >) 丨. The covariance of 匕(4) can be given as zero = μ n. The sinr of the /th element can be expressed as: SINR {bt^(s)} =-ΙΑί^ρΐ2 Σμ,αοι2

Equation (52) where .rLcd G is the (7, /) element of b, and ^•WW, C (X 〇 is Kw W, and the number of C (7, element. SINRP/.^)} is borrowed The SINR of the first data stream sent by the channelization code c, and can be used to select the data transmission rate for the data stream. The SINR of each channelization code c depends on the transmission matrix used for the channelization code. Receiver 150 determines the sSINR for different possible transmit matrices and selects the transmit matrix with the highest SINR. Receiver 15A can send feedback information to transmitter 110. This feedback information can include the selected for each channelization code. Transmit matrix, SINR or data transmission rate per channelization code, average SINR or data transmission rate of all channelization codes, etc. 116763.doc -29· 1345895 In general, front-end filtering can be used in the first Li signal. The on-time signal component.,*, 'and the middle of the processing to process the M receivers before the transmission is processed at the transmitter end-like filter - the non-view signal is applied to the front-end filter for all 诵^ for CDM transmission, Number component to recover L transmissions: No. 2 Alignment time signal visible signal before the launch wn thief ~ social type (for example, transmitter transmitted matrix I and gain material (6) and complex (four) used ', factors (for example, channel back; ® g and signal The statistics are up to „). Difficult, W Li Η Figure 5 shows a design of the process of recovering the box _ ΜΙΜΟ transmission at the receiver. It is obtained for processing (for example, ritual, suppression or mitigation) The front-end ferrocoupler of the non-punctual signal component of the received signal (block 512). The front-end chopper does not isolate the non-punctual signal component. In fact, the front-end chopper processes the non-punctual signal component in a desired/beneficial manner. And (also incidentally) processing the on-time signal component. Also resulting in at least a combiner matrix for combining the on-time signal components for the plurality of transmitted signals (block 514). The front-end ferrite can be as shown in equation (32) And the resulting matrix, and the combiner matrix can be derived as shown in equation (33). The front-end filter can also be obtained as shown in equation (26) or (28); And the combiner matrix can be (for example As shown in equation (43), (47) or (50); ^. The front-end filter and combiner matrix can also be derived in other ways. It can be based on the received data for the pilot and according to (for example The front-end filter is derived from the least squares criterion. The combiner matrix can be derived based on the transmit matrix used to transmit the data, the gain for the data, the channel response estimate, the front-end filter, the signal and/or the noise statistics, and the like. The combiner matrix can also be derived from MMSE or some other standard. 116763.doc -30- 1345895 1345895

The received resource_waves are processed to process a plurality of received signal median components (block 516). The punctual signal component is combined (block 518) by the processing of the signal (4). The received data may be more than one symbol period to process the non-punctual signal component 'via the wave: can be processed for the duration of the symbol period to combine the on-time signal components. The information received and the data obtained by chopping can be given in the form of samples, symbols, etc. ...... Pair: With CDM transmissions sent by multiple channelization codes, an early-end filter can be derived and used to process non-punctual signal components, and a combined state matrix can be derived and Used to combine on-time signal components for multiple channelization codes. In the mechanism, the money is first obtained by the former wave device to obtain the intermediate data. The intermediate data is then despread for each channelization code to obtain the data for the channelization code. For each per-channelization code, the output data for the channelization code is obtained by a matrix of the combiner matrix for the channelization code. In another mechanism, the received data is first despread for each channelized code to obtain the spread spectrum data for the channelized code. The same-% filter filters the despread spread data for each channelization code to obtain filtered data for the channelization code. For each pass-through code, the filtered 'wave' data is further processed by a combiner matrix for the channelized code to obtain output data for the channelized code. Sub-CDM and may be based on received data and known pilots, such as based on () received data for chip level training and samples of known pilot chips or (b) self-signed symbol level The front-end filter is obtained by training the received data and the despread pilot symbols obtained by the known pilot symbols. Can be based on the multi-channelization code for the multiple 116763.doc -31 · channelization code of the spread matrix matrix 1 in the multi-channelization code gain 'channel response estimation, front-end filter, 1 port and / or noise statistics Or combine to get the combiner matrix.前娅冶.士 „ Λ 八,. Estimated (for example) as wpwg. τ j gossip

Multi-level connection ◎ can also be used for other communication systems. For example, in time division multiplexing (in the TD system, the terminal filter can be paid based on the pilot received in the first time interval) and can be based on the transmission matrix used in the second time interval. The combiner matrix of the second time interval of the poem is obtained. The data received in the second time interval can be filtered by the front end filter, and the chopped data can be further processed by the combiner matrix. 'Filters can be derived based on pilots, which can be transmitted at a particular channelization code and/or time interval and using a particular transmit matrix and gain. The pilot derived filter can be used to derive a data filter, It can be transmitted at other channelization codes and/or time intervals and possibly with different transmit matrices and gains. • In CDMA, 'punctual and non-punctual signal components can be distinguished by their transmit time. The receiver can handle a sample window to recover the desired symbol transmitted by the transmitter. The timing of the equalizer determines the time instant of transmitting the desired symbol relative to the window. The sample obtained by the receiver includes punctuality and Different extra signal components on time 彳 s component. The punctual signal component is a signal 3 for the desired symbol and other symbols transmitted simultaneously with the desired symbol. All other signal components are non-punctual signal components, including tracking to The signal component of the symbol transmitted before and after the symbol. The symbol can be transmitted by the transmit function, which may depend on one or more parameters. 116763.doc •32· For example, S, the transmit function can be a miscellaneous girl. Cycle · 5, channelization code C, frequency bin or subcarrier index "etc. and the table is not. For the sake of simplicity, the 'emission function can be seen from three parameters ~ and", or tuple (four). = The emission function of different symbols can be orthogonal such that only ❿, 1~"2</(51,">1,~), 2, and 2)> Expressed as (5l, Cl, Wl) = (*s2, c2, /i2).彳m enclosed (4) from what is defined by the tuple heart 1, "1):: function plus .... the desired signal component and (9) from other transmit letters ~ where (the other signal components of the object. first-level wide-end German The wave can process other signal components. The combiner in the second stage can process the desired signal component. For CDM, the transmit function for the symbol period is determined by the (four) Μ fine (four) order of length C. It is used for the symbol period recognition channel. The emission function of the code c can be expressed as ^ ^ 八) From the point of view of the emission function / (he丨) emission, the received signal contains the following: 1. should be in /〇^) (shirt heart) Non-punctual signal component, 2mc) and includes the following on-time signal components: a. The required channelized code from the desired channelization code is added, ι), and b, from other channelization codes. Called the on-time signal component 0 The front-end filter processes the non-punctual signal component corresponding to /(,,c). The descrambling and despreading by the front =' filter also eliminates the other materialized codes and corresponds to Take the on-time signal component of Cl). The combiner processes the code from the desired channel and corresponds to /( , 〇) The on-time signal component 116763.doc •33- 1345895 In a single-carrier system that does not use reading, the emission function can only be a digit Δ that varies with time and can be given as eight (four) (four). Over time, over time, The position of Δ changes with time.

Based on. The 'emission function' can be used for different subcarriers in the system and can be given as /(u) ' where # is the subcarrier index. The subcarrier in 〇fdm may correspond to the channelization code in (10). The transmitter may convert (N) data/pilot symbols into time domain by the inverse fast Fourier transform (IFFT) to obtain N time domain samples and (b) append a cyclic preamble to the time domain samples by: An OFDM symbol is derived and the given transmit antenna transmits the material/pilot symbols on the N subcarriers in the symbol period. The receiver may (4) remove the cyclic preamble in the received samples and (b) convert the N received samples into the frequency domain by fast Fourier transform (FFT) to obtain N receptions of the serving subcarriers. The received data/pilot symbols for the given receive antenna are obtained by the symbol. The received symbol may correspond to 匕 (4, where subscript C is replaced by subcarrier index „. For 〇FDM, the punctual signal component may be a signal transmitted from a different transmit antenna on a subcarrier. The non-punctual signal component can be a signal component transmitted on other subcarriers. The front end chopper can be implemented by removing the FFT and the cyclic preamble at the receiver. The combiner matrix can be done for each subcarrier. Calculated and used to combine received symbols from all receive antennas for the subcarrier. Those skilled in the art will appreciate that information and signals can be represented using any of a variety of different techniques and techniques. The materials, instructions, commands, information, signals, bits, symbols and chips cited in the above description may be represented by voltage, current, electromagnetic waves, magnetic fields or particles, optical fields or particles or any combination of 116763.doc • 34· Those skilled in the art will further appreciate that the various illustrative logic blocks, modules, circuits, and algorithm steps described in connection with the disclosure herein can be implemented as electronically hard. , computer software, or a combination of both. To clearly illustrate this interchangeability of hardware and software, various illustrative components, blocks, modules, circuits, and steps have been described above generally in terms of their functionality. The functional implementation is hardware or software depending on the particular application and the constraints imposed on the overall system. Those skilled in the art can implement the described functionality in different ways for each particular application, but Such implementation decisions should not be interpreted as causing a departure from the scope of the present disclosure. The various illustrative logic blocks, modules, and circuits described in connection with the disclosure herein can be implemented as a general purpose processor designed to perform the functions described herein. , digital signal processor (DSP), special application integrated circuit (ASIC), field programmable gate array (FPGA) or other programmable logic device, discrete idle or transistor logic, discrete hardware components or any of them - The general purpose processor may be a microprocessor 'but in an alternative embodiment, the processor may be any conventional processor, controller, microcontroller or state The processor can also be implemented as a combination of computing devices, such as a combination of a Dsp and a microprocessor, a combination of a plurality of microprocessors n, together with -Dsp(4)- or a combination of multiple microprocessors or any other such configuration. The steps of the method or algorithm described in connection with the disclosure herein may be implemented directly in hardware, in a software module executed by a processor, or in a combination of the two. The software module may reside in RAM memory, flashing Memory, R0M^ memory, EPR (10) memory, EEP memory, scratchpad, hard disk, 116763.doc • 35· 1345895 removable disk, CD-ROM or any other form known in the art In the storage medium, the exemplary storage medium is coupled to the processor such that the processor can read the information from the storage medium and write the information to the storage medium. In an alternative embodiment, the storage medium can be integrated into the processor. The processor and storage media can be resident in the ASIC. The ASIC can be resident in the user terminal. In an alternate embodiment, the processor and storage medium may reside as discrete components resident in the user terminal.

The previous description of the disclosure is provided to enable any person skilled in the art to make or use the disclosure. Various modifications to the present disclosure will be readily apparent to those skilled in the art, and the general principles defined herein may be applied to other variations without departing from the spirit or scope of the disclosure. Therefore, the present disclosure is not intended to be limited to the examples described herein, but is consistent with the broadest scope of the principles and novel features disclosed herein. BRIEF DESCRIPTION OF THE DRAWINGS Figure 1 shows a block diagram of a transmitter and a receiver.

Figure 2 illustrates a MIMO-CDM transmission. Figure 3 shows a block diagram of a CDMA modulator. Figure 4 shows the design of the various blocks within the receiver. Figure 5 shows the process of recovering a transmission at the receiver. [Main component symbol description] 110 Transmitter 112 Transmit data processor (τχ data processor) 114 ΤΧ ΜΙΜΟ processor 116 CDMA modulator 116763.doc -36- 1345895

116a-11 6m CDMA modulator 118a-11 8m transmitter unit 120a-120m antenna 130 controller/processor 132 memory 150 receiver 152a-152n antenna 154a-154n receiver unit (RCVR) 156 channel processor 160 equalization / CDMA Demodulator / Block 170 Receive (RX) MIMO Processor / Block 172 RX Data Processor 180 Controller / Processor 182 Memory 310 Data Processor 312 Spread Spectrum 314 Multiplier 320 Pilot Processor 322 Spreader 324 Multiplier 330 Summer 332 Scrambler 410 Front End Filter/Equalizer 420 CDMA Demodulator 116763.doc -37- 1345895 422a-422c De-scrambler/De-spreader 432a-432c Combination 442 timing estimator 444 coefficient calculation unit 446 ττ〇 early 兀 448 channel estimator 450 early 兀 452 στ» — early 兀 116763.doc -38-

Claims (1)

1345895 Patent No. 0951 <44483 Chinese application for the patent garden replacement (in February of the following year) Γ" X. Patent application park: 1. 0 years 2. Moon • 1. A device for wireless communication, including: 'at least one processor Filtering the received data to process non-punctual signal components of the plurality of received signals and obtaining filtered data 'and processing the filtered data to combine the on-time signal components for the plurality of transmitted signals, At least one processor estimating a quality of the received signal for the at least one data signal transmitted in the plurality of transmitted signals; and The 5 memory is attached to the at least one processor. The apparatus of claim 1, wherein the at least one processor filters the received tributary for more than one symbol period to process the non-punctual signal components, and processes the filtered data for one symbol period Combine the on-time signal components. 3' If the device is requested, the at least the processor filters the received data in time. 4. The apparatus of claim 1 wherein the at least one processor derives a front end filter for processing the non-punctual signal components and derives at least a combiner matrix for combining the on-time signal components. 5. The apparatus of claim 4 wherein the at least one processor derives the front end filter based on the received data for the pilot and based on at least one of transmitting data in the plurality of transmitted signals The matrix is emitted to derive the at least one combiner matrix. 6. The apparatus of claim 1, wherein the at least one waver is configured to process the non-punctual signal sub-processor to derive a front-end filter quantity' and to derive a plurality of combinations 116763-1000201.doc 1345895 matrix for use The on-time signal components are combined for a plurality of channelization codes for the plurality of transmitted signals. The device of claim 6, wherein the at least one processor filters the received data by the front end filter, and the filtered data is processed by the plurality of combiner matrices to obtain the plurality of The output data of the channelization code. The device of item 6, wherein the at least one processor filters the received data by the front end chopper to obtain intermediate data, and the intermediate data is performed for each of the channelization horses Despreading the frequency to obtain the chopped data for the channelization code, and processing the filtered data for each channelized code by using the channelizer matrix for the channelization to obtain The output data of the channelization code. 9. The device of claim 6, wherein the processor of _.u, " to v is for each of the plurality of passivation codes to perform the spread of the sigh The demodulation frequency for the 4-channelization code is used for the mother-channelization code to obtain the data for the channelization code by using the spread spectrum, and for each - if servant 4·<4<> The output data for the channelized code is filtered by a combiner matrix for the channelization code. $仃Processing to obtain 10. As claimed in claim 6 The device, wherein the processor is determined to derive the front-end chopper based on the received subject and the known pilot. #接收U· The device of claim 6, wherein the at least - The received data and the known pilot memory processor are based on the device for use. The front end filter I16763-I〇〇〇2〇ld〇c 1345895 is obtained. The device of claim 6, wherein the at least A processor despreads the data received by the pilot by a pilot channelization code to obtain a despread pilot number The front-end filter is obtained by de-spreading the pilot symbols and the known pilot symbols. For example, the device of the monthly solution 6 wherein the at least one processor derives the front-end filter based on the least squares criterion. The apparatus of claim 6, wherein the at least one processor derives the plurality of combiner matrices based on a plurality of transmit matrices for the plurality of channelization codes. 15. The apparatus of claim 14, wherein The at least one processor further derives the plurality of combiner matrices based on the gain of the evening channelization code. 16_ The apparatus of claim 14 wherein the at least-processor is further based on a channel response estimate and the front end filter The plurality of combiner matrices are derived. The apparatus of claim 14, wherein the at least one processor derives the plurality of combiner matrices based on a minimum mean square error (MMSE) criterion. The apparatus of claim 6, wherein the at least one processor is based on a noise covariance matrix, the front end filter, a channel response estimate, and a transmit matrix for the channelization code for each channelization code. A combiner matrix is produced. 19. The apparatus of claim 6, wherein the at least one processor derives a correlation matrix based on the filtered data, and for each channelization code based on the correlation matrix, the front end filter, a channel estimate, and a A combiner matrix is obtained for the channelized mother's emission matrix. 20. The apparatus of claim 6 wherein the at least one processor is updated by a first update 116763-1000201.doc 21 21 and at a different matrix than the first update rate. Rate updating the front end filter, and the second update rate is updated by the dream device of the claim 1 wherein the at least one processor is derived based on the pilot received in a first time gate interval - the front end a waver, based on the emission matrix for the first time interval, to obtain a combination matrix for the second time interval, wherein the front end filter pairs the received data for the second time interval (4) The wave obtains the chopped data for the second time interval and the filtered data is processed by the combiner matrix. 22. An apparatus for wireless communication, comprising: to a processor that derives a front end filter for processing non-punctual signal components of a plurality of received signals to derive a plurality of combiner matrices And combining a plurality of transmitted signals transmitted by the plurality of channelization codes to combine the on-time signal components, wherein the received samples are filtered by the front-end filter and the filtered signals for the plurality of channelization codes are obtained. And obtaining, for each of the plurality of channelization codes, an output symbol for the channelization code by a combiner matrix for the channelization code, wherein the at least one processor is for the plurality of transmissions At least one data signal transmitted in the signal estimates the quality of the received signal; and a memory element is consuming to the at least one processor. 23. The apparatus of claim 22, wherein the at least one processor derives the front end filter based on the received samples and known pilot chips. 24. The apparatus of claim 22, wherein the at least one processor derives a combiner I16763-I000201.doc -4·1345895 matrix for each channelization code based on a transmit matrix for the channelization. A method for wireless communication, comprising: filtering received data to process non-punctual signal components of a plurality of received signals and obtaining filtered data; processing the filtered data for a plurality of The transmitted signal combines the on-time signal component; and estimates the quality of the received signal for at least one of the data signals transmitted in the plurality of transmitted signals. The method of claim 25, further comprising: deriving a front end filter for processing the non-punctual signal components; and ~ extracting at least one combiner matrix for combining the on-time signals Component. 27. The method of claim 25, further comprising: deriving a threshold filter for processing the non-punctual signal components; and ° '. A plurality of combiner matrices are derived for combining the on-time signal components for a plurality of channelization codes for the plurality of transmitted signals. 28. The method of claim 27, wherein the received data is chopped by the end m by filtering the received data to obtain intermediate data 'and for the plurality of channels Each of the intermediate data is despread to obtain a filtered material for the channelization code, and the filtered data is included for each channel. 116763-I00020I.doc is called The filtered data is processed by a combiner matrix for the channelized code to obtain output data for the channelized code. 29. The method of claim 27, wherein the deriving the front end filter comprises deriving the front end filter 基于 based on samples for the received data and known pilot chips. 3. The method of claim 27, wherein the deriving the plurality of combiner matrices comprises deriving a combiner for each of the plurality of channelized codes based on a transmit matrix for the channelized code matrix. 3l. Apparatus for wireless communication, comprising: means for filtering received data to process non-punctual signal components of a plurality of received signals and obtaining filtered data; The filtered data is a component that combines the on-time signal components for the plurality of transmitted signals; and means for estimating the quality of the received signal for the at least one data signal transmitted in the plurality of transmitted signals. 32. The apparatus of claim 31, further comprising: means for deriving a front end filter for processing the non-punctual signal components; and for deriving a plurality of combiner matrices for use in A plurality of channelization codes of the plurality of transmitted signals combine components of the on-time signal components. Means of the monthly claim 32, wherein the means for performing the wave on the received data includes means for filtering the received data by the front end filter to obtain intermediate data and using De-spreading the intermediate data for each of the plurality of quantization codes to obtain a component for filtering the data of the channelization code with 116763-1000201.doc 1345895 and for processing the filtered And the component of the data of the wave includes processing, for each channelization code, the data of the (4) wave of the combiner matrix for the channelization code to obtain the output data for the channelization code. The device of claim 32, in the basin, the abbreviated a, the animal 1 , the component used for the front-end chopper includes, for the purpose of receiving the received data and the known pilot chip Deriving a component of the front end money, and wherein the means for deriving the plurality of composite materials comprises for transmitting, for each of the multi-channelization codes, a transmission for the channelization code The matrix yields the components of a combiner matrix. 35. A processor readable medium for storing instructions operable to: chop the received data to process a plurality of received non-punctual signal components and obtain filtered data; U processing „ The nucleus of the eclipse combines the on-time signal component with a plurality of transmitted signals; and ▲ estimates the quality of the received signal for at least the data signal transmitted in the four transmitted signals. An apparatus for wireless communication, comprising: at least one processor that performs processing of non-punctual signal components of a plurality of received signals to obtain received symbols, resulting in a combiner=array for on-subcarriers The plurality of transmitted signal groups transmitted on time are on time L-number, and the received symbols for the sub-carrier are processed by the combiner matrix to obtain an output symbol for the sub-carriers 116763-1000201.doc 丄j j j j j j j j j j j j j j j j j j j j j j j j j j j j j j j j j j j j j j j j j j j j j j j j j j j j j j j j j j j j j j j j j j j j j j j j j j j j j j j j j j j j j j j j j j j j j j j j The to A 37. In the case of claim 36, Ganzi ^ ', the at least one processor is removed by removing the pre-term and the received 1" by removing the broadcast ring Fast Fourier Transform (FFT) And modulating the received symbols and performing the non-punctual signal components: the apparatus of claim 36 wherein the at least one processor is based on a noise co-matrix, a channel response estimate, and a transmission moment for the subcarrier The combiner matrix is derived from at least one of 11. The apparatus of claim 36, wherein the at least one processor derives a second set of betator matrices for combining a plurality of transmit apostrophes to transmit a punctual signal component on a second subcarrier, and by using the A second combiner matrix processes the received symbols for the second subcarrier to obtain an output symbol for the second subcarrier. 116763-1000201.doc 8-