HK1064525B - Single user detection - Google Patents

Description

Single user detection

Background

This application claims priority to U.S. provisional patent application 60/246947, filed on.11/9/2000.

The present invention relates to wireless communication systems. In particular, the present invention relates to data detection in wireless communication systems.

Fig. 1 illustrates a wireless communication system 10. The communication system 10 has a base station 12₁To 12₅(12) Which is associated with a User Equipment (UE)14₁To 14₃(14) Communication is performed. Each base station 12 has an associated operating region, wherein the base station 12 communicates with User Equipment (UE)14 in its operating region.

In some communication systems, such as: code Division Multiple Access (CDMA) and time division duplex (TDD/CDMA) using code division multiple access, multiple communications are transmitted over the same frequency spectrum. These communications are distinguished by their channel coding. To more efficiently use this spectrum, time division duplex (TDD/CDMA) communication systems utilizing code division multiple access use repeating frames that are further divided into a plurality of communication slots. A communication transmitted in such a system is assigned one or more associated information codes and time slots. The use of an information code in a slot is referred to as a resource unit.

Since multiple communications may be transmitted over the same frequency spectrum and at the same time, a receiver in such systems must be able to distinguish between the multiple communications. One method of detecting such signals is multiple user detection. In multi-user detection, the signals of the users are detected simultaneously in relation to all User Equipments (UEs) 14. A method of performing multi-user detection includes block linear equalization based joint detection (BLE-JD) using a Cholesky or an approximate Cholesky decomposition. These methods have a high degree of complexity. This high complexity increases power consumption and is experienced at the User Equipment (UE)14₁Further reducing battery life. It is therefore a primary object of the present invention to provide other methods for detecting received data.

Summary of The Invention

A transmitter side transmits a plurality of data signals over a shared bandwidth of a Code Division Multiple Access (CDMA) communication system. Wherein each transmitted data signal experiences a similar channel response. First, a combined signal of the transmission data signals is received. The combined signal is then sampled using a multiple of the chip rate. Then, a channel response of the combined signal is determined. Then, a first element of a spread data vector is determined using the combined signal samples and the estimated channel response. Then, using a factor of the first element determining step, the remaining elements of the spread data vector are determined. Finally, the data of the data signal is determined using the determined elements of the spread data vector.

Drawings

Fig. 1 is a wireless communication system.

Fig. 2 is a simplified transmitter and a single user detection receiver.

Fig. 3 is a communication burst (burst).

FIG. 4 is a flow chart of an extended forward alternative to Single User Detection (SUD).

FIG. 5 is a flow chart of a approximate stripe Cholesky method of Single User Detection (SUD).

FIG. 6 is a flow chart of a Toeplitz method of Single User Detection (SUD).

FIG. 7 is a flow chart of Fast Fourier Transform (FFT) applied to the channel correlation matrix for Single User Detection (SUD).

FIG. 8 is a flow chart of a Fast Fourier Transform (FFT) using efficient combinatorial Single User Detection (SUD).

FIG. 9 is a flow chart of a Fast Fourier Transform (FFT) of a Single User Detection (SUD) filled with zeros.

Detailed Description

Fig. 2 shows a simplified transmitter 26 and receiver 28 using Single User Detection (SUD) in a time division duplex (TDD/CDMA) communication system using code division multiple access, although the single user detection method can be used in other systems, such as: frequency division duplex (FDD/CDMA) using code division multiple access. In a typical system, a transmitter 26 is located in each User Equipment (UE)14 and a plurality of transmission circuits 26 for transmitting a plurality of communications are located in each base station 12. The Single User Detection (SUD) receiver 28 may be located at a base station 12, at a User Equipment (UE)14, or both. Typically, Single User Detection (SUD) detects data transmitted via a particular transmitter in a single or multiple information code (multicode) transmission. Each channel information code signal (channel code signal) in such a multiple information code transmission experiences the same channel impulse response (channel impulse response) when all signals are transmitted via the same transmitter. In particular, Single User Detection (SUD) is used for downlink transmissions, where all transmissions originate from a base station antenna or antenna array. In addition, Single User Detection (SUD) may also be used for uplink transmissions, where a single user transmits exclusively a timeslot using a single or multiple information codes.

The transmitter 26 transmits data over a wireless radio channel (wireless radio channel) 30. A data generator 32 in the transmitter 26 generates data to be transmitted to the receiver 28. A modulation/spreading sequence insertion device 34 generates a communication burst (burst) or bursts by spreading the data using a midamble in the appropriately assigned time slot and spreading the data with information codes and time-division multiplexing the spread reference sequence data.

A typical communication burst 16 has a midamble 20, a guard period 18, and two data bursts 22, 24, as shown in fig. 3. The text 20 separates the two data fields 22, 24 and the guard period 18 separates the communication bursts to allow for time differences in arrival of the transmission bursts of different transmitters 26. The two data bursts 22, 24 contain data for the communication burst.

The communication burst or bursts are modulated to Radio Frequencies (RF) using a modulator 36. An antenna 38 transmits the Radio Frequency (RF) signal through the wireless transmission channel 30 to an antenna 40 of the receiver 28. The type of modulation used for the transmission communication may be of any type familiar to those skilled in the art, such as: quadrature Phase Shift Keying (QPSK) or M-ary Quadrature Amplitude Modulation (QAM).

The antenna 40 of the receiver 28 receives various Radio Frequency (RF) signals. These received signals are demodulated by a demodulator 42 to generate a baseband signal. The baseband signal is sampled using a sampling device 43, such as one or more analog-to-digital converters, using the chip rate of the transmission bursts or a multiple of the chip rate. The samples are processed, for example, using a channel estimation device 44 and a Single User Detection (SUD) device 46, in the time slot using the appropriate information codes assigned to the received bursts. The channel estimation device 44 uses the midamble elements of the baseband samples to provide channel information, such as: the channel impulse response. This channel impulse response can be seen as a matrix H. The channel information is provided to the Single User Detection (SUD) device 46 for use in estimating the transmitted data of the received communication bursts for use as soft symbols (soft symbols).

The Single User Detection (SUD) device 46 uses the channel information provided by the channel estimation device 44 and the known spreading information code used by the transmitter 26 to estimate the data of the desired received communication bursts. Although the Single User Detection (SUD) is described using the 3GPP (third generation partnership project) Universal Terrestrial Radio Access (UTRA) Time Division Duplex (TDD) system as the base communication system, the Single User Detection (SUD) can be applied to other systems. The system is a direct sequence wideband code division multiple access (W-CDMA) system in which uplink and downlink transmissions are restricted to mutually exclusive time slots.

The receiver 28 receives a total of K bursts 48 that arrive simultaneously using its antenna 40. The K bursts overlap each other during a single observation. For third generation partnership project (3GPP) Universal Terrestrial Radio Access (UTRA) Time Division Duplex (TDD) systems, each data segment of a timeslot corresponds to a single observation period.

For an observation period, this data detection problem can be known from equation (1):

r ═ H · d + n equation (1)

Where r is the received samples. H is the channel response matrix. d is the spread data vector. The spread data matrix includes data for each channel transmission that is mixed with the spreading information code for that channel.

When the received signal is oversampled, multiple samples of each transmitted chip are generated, thereby obtaining a received vector r₁，r₂，…，r_N(48). Similarly, the channel estimation device 44 determines the channel responses H₁、H₂、…、H_NCorresponding to these received vectors r₁，r₂，…，r_N(50). For a double chip rate, equation (1) will evolve into equation (2):

equation (2)

Wherein r is₁These even samples (using the chip rate), and r₂The odd samples (and r₁Sample offset by half chip). H₁Is the channel response matrix of these even samples, and H₂The odd sampled channel response matrices.

For N chip rates, equation (1) will evolve into equation (3):

equation (3)

Wherein r is₁、r₂…r_NIs a multiple of these chip rate samples, where each offset is 1/N chip. H₁、H₂、…、H_NCorresponding to the channel response. Although the following discussion focuses on a receiver sampling at twice the chip rate, the same method can be applied to any multiple of the chip rate.

For double chip rate sampling, the matrix H₁And H₂The size of which is (N)_S+W-1)×N_S。N_SThe number of spreading chips transmitted during the observation period, and W is the length of the channel impulse response, such as: a chip of 57 chips. Since the received signal has N_SIndividual spreading chip code, r₁And r₂The length of which is N_S. Thus, equation (2) will be rewritten as equation (4):

equation (4)

Wherein r is₁(i)、r₂(i)、h₁(i)、h₂(i) Are respectively the corresponding vector matrixes r₁、r₂、H₁、H₂The ith element of (a).

One method of determining the spread data vector is the extended forward substitution method, the steps of which are described in detail in conjunction with fig. 4. For the extended forward substitution method, the received data vector is rearranged such that each even sample follows its corresponding odd sample. In addition, a similar rearrangement is performed on the channel response matrix, as shown in equation (5 a):

equation (5a)

Similarly, for N times chip rate sampling, the arrangement is shown in equation (5 b):

equation (5b)

d (i) is the i-th element of the spread data vector d. The length of the spread data vector is N_S. The zero-forcing solution for determining d (0), d ^ 0, by using the extended forward substitution method is based on equation (6a) and equation (7 a): (52)

equation (6a)

Equation (7a)

Equation (6a) is a general formula for d (0). Equation (7a) is zero-forcing solution for d ^ (0). Similarly, for N times the chip rate, equations (6b) and (7b) can be used:

equation (6b)

Equation (7b)

In solving equations (7a) and (7b), v is used to provide subsequent operations^HIs determined by using equation (8) to obtain v of equation (7a)^HAnd storing: (52)

equation (8)

And d ^ (0) is v^HDetermined according to equation (9):

equation (9)

Using the Toplitz structure of the H matrix, the remaining spread data elements can be sequentially determined using zero forcing (zero forcing) according to equation (10 a): (54)

equation (10a)

For N chip rates, equation (10b) can be used:

equation (10b)

After the spread data vector is determined, the data for each communication burst is determined using a despreading method, such as: this spread data vector is mixed with the information code of each burst. (56)

The complexity of using this extended forward substitution method (excluding the despreading method) is summarized in the first table.

Calculating V	Four multiplications and one reciprocal
		Computing d ^ (0)	Two multiplications
Computing d ^ (1)	Four multiplications
		Calculating each solution until d ^ (W-1)	Two multiplications
From d ^ (W) to d ^ (N)-1) calculate each d ^ (i)	(2W +2) multiplications
		Total number of multiplications	2N+(W-1)W+2W..(N-W+1)
Calculating the total number	2N+(W-1)W+2W..(N-W+1)+5

TABLE 1

For a Time Division Duplex (TDD) burst type II, N_SIs 1104 and W is 57, with 200 times per second extended forward substitution, the solution d requires 99.9016 million real-time Operations Per Second (MROPs) (for double chip rate sampling) or 49.95 million real-time Operations Per Second (MROPs) (for chip rate sampling).

Another method of estimating the data is the approximate stripe Cholesky method, the steps of which are described in detail below with reference to FIG. 5. Wherein a cross-correlation matrix R is determined so that it becomes a square (N)_S×N_S) And becomes a stripe according to equation (11): (58)

R＝H^Hh equation (11)

Wherein, ()^HIt represents the Hermetian function. H is 2 (N)_S+W-1)×N_S. For double chip rate sampling, equation (11) is rewritten as equation (12 a):

equation (12a)

Alternatively, for N times chip rate sampling, equation (12b) may be used.

Or

Equation (12b)

Using either equation (12a) or (12b), for double chip rate sampling, the cross-correlation matrix R is obtained to have a size N_S×N_SAnd becomes a stripe according to equation (13), where W equals 3 and N_SEqual to 10.

Equation (13)

Generally, the bandwidth of the cross-correlation matrix R is determined according to equation (14):

w-1 equation (14)

Cross-correlating sub-blocks of a matrix R (R) using an approximate Cholesky method_SUB) The size of which is N_col×N_colCan be used. This sub-block R_SUBThe typical size is (2W-1) × (2W-1), although other sizes of matrix may be used. This sub-block (R)_SUB) Decomposed using Cholesky decomposition according to equation (15): (60)

R_sub＝GG^Hequation (15)

The size of the Cholesky factor G is N_col×N_col. A 5 × 5 Cholesky factor G matrix (W ═ 3) is determined according to equation (16):

equation (16)

Wherein G is_ijIs the element of the Cholesky factor G matrix in the ith column and jth row. The Cholesky factor G matrix is extended to N by right-shifting the last row of the Cholesky factor G matrix by one element after the last row of the Cholesky factor G matrix_S×N_SMatrix G of_full(62). For N_SEquation (16) is extended according to equation (17) for 10. (62)

Equation (17)

The spread data vector is determined using forward and backward substitution (64). For double chip rate sampling, the forward substitution method determines y according to equation (18 a); also, for N chip rate sampling, the forward substitution method determines y according to equation (18 a).

Equation (18a)

Equation (18b)

The backward substitution method then solves for this spread data vector according to equation (19).

Equation (19)

After the spread data vector (d) is determined, the individual burst data is determined (66) using a despreading method.

For double chip rate sampling, the complexity of the approximate Cholesky decomposition method (excluding the despreading method) is determined according to a second table.

Operations	Number of calculations
		Calculate HH	W(W+1)

Computational Cholesky decomposition method	N(W-1)/2+3N(W-1)/2-(W-1)/3-(W-1)-2(W-1)/3
		Calculate Hr	2NW
Forward substitution method	〔N- (W-1)/2) W and NReciprocal of real number
		Backward substitution method	〔N- (W-1)/2) W and NReciprocal of real number

TABLE 2

For a Time Division Duplex (TDD) burst type II, N_S1104, and for W57, performing the approximate stripe Cholesky method 200 times per second with twice the chip rate requires 272.56 million real time operations per second (MROPS). In contrast, a precision stripe Cholesky approach requires 906.92 million real time operations per second (MROPS). For chip rate sampling, the approximate stripe Cholesky method requires 221.5 million real time operations per second (MROPS).

Another method of data detection uses a Toeplitz method (Levinson-Durbin type algorithm), the steps of which are described in detail below in conjunction with FIG. 6. Here, the cross-correlation matrix R of equations (12a) and (12b) is reproduced according to equations (12a) and (12 b):

equation (12a)

For N chip rates, equation (12b) can be used:

or

Equation (12b)

The cross-correlation matrix R is symmetric and Toeplitz (bandwidth p ═ W-1) (68). The upper left corner R (k) of the cross-correlation matrix R, which is a k matrix, is determined according to equation (20):

equation (20)

In addition, another vector R_kIs determined according to equation (21) using the elements of the cross-correlation matrix R: (72)

equation (21)

The bold system represents a matrix that includes all elements up to its subscript. At order (k +1), this system solves according to equation (22):

R(k+1)d(k+1)＝[H^Hr]_k+1equation (22)

[H^Hr]_k+1Is H^Hr is preceded by (k +1) components. d (k +1) is decomposed into a vector d of length k according to equation (23)₁(k +1) and scalar d₂(k+1)：

Equation (23)

This matrix R (k +1) is decomposed according to equation (24):

equation (24)

E_kIs a switching matrix. The switching matrix inverts all elements of a vector.

Using Yule-Walker equation for linear estimation, equation (25) yields (78):

equation (25)

Using sequential recursion, equations (26), (27), and (28) can be derived:

y₁(k)＝y(k-1)+y₂(k)E_k-1y (k-1) equation (26)

Equation (27)

Equation (28)

With y (k), d (k +1) is determined (74) according to equation (29), equation (30), and equation (31):

d₁(k+1)＝d(k)+d₂(k+1)E_ky (k) equation (29)

Equation (30)

Equation (31)

Wherein (H)^Hr)_k+1Is H^HThe (k +1) th element of r.

After appropriate initiation of these loops, they are directed to k 1, 2_SAnd (6) performing calculation. d (N)_S) Is the solution (74) of equation (32):

Rd＝H^Hr equation (32)

The spread data vector d is encoded with a burst of channel information to recover data (76).

The fringe structure of the cross-correlation matrix R affects these recursions as follows. R (2) and R2 are determined according to equation (33):

equation (33)

The inner product calculations in equations (27) and (30) require two multiplications, respectively. For convenience of explanation, the cross-correlation matrix R (k ═ 6 for example) is determined according to equation (34):

equation (34)

This vector R₆The number of non-zero elements is equal to the bandwidth p of the cross-correlation matrix R. When the inner product R in equation (27) is calculated₆ ^HE₆y (k) and the inner product R in equation (30)₆ ^HE₆d (k), this method requires only p (rather than k) multiplications. For the recursion of equations (26) and (29), this approach does not reduce any computation.

The third table represents the complexity of implementing the Toeplitz method.

Computing	Number of calculations	Million instant operations/second (MROPS)
			Each burst calculates HFunction executed once in H time		1.3224
Solve Yule-Walker to get y	672888×100/10	269.1552

Each burst calculates Hr is twice executed		100.68
			Answer R (k +1) d (k +1) Hr	672888×200/10	538.3104

TABLE 3

For a Time Division Duplex (TDD) burst type, the total million real-time operations per second (MROPS) of this Toeplitz method is 909.4656 million real-time operations per second (MROPS) (using two chip rate sampling), and 858.4668 million real-time operations per second (MROPS) (using one chip rate sampling).

Another method of data detection uses a Fast Fourier Transform (FFT), the steps of which are described in detail below in conjunction with FIG. 7. If sampling is performed at a chip rate, the channel matrix H is square, except for edge effects (edge effects). Using a cyclic approximation of the H matrix, data estimates are obtained using a Fast Fourier Transform (FFT) of the received vector and the channel vector H.

For multiple chip rate sampling, such as double chip rate sampling, this H matrix is not square and cyclic. However, the channel correlation matrix R of equation (13) is H^HA submatrix of H, shown as a dashed line, is cyclic, as shown in equation (35 a):

equation (35a)

For an N times chip rate sampling, the channel correlation matrix is determined according to equation (35 b).

Equation (35b)

Approximating the channel correlation matrix R as a circular, equations (36), (37), and (38) can be used:

R^H＝DΔD^Hequation (36)

Wherein Δ is equal to diag (D (R))₁)

Equation (37)

Wherein, (R)₁The first column of the correlation matrix R extends as a diagonal matrix (diagonalmatrix). Although the first row is illustrated, the method can be adapted to be applied to any row 86 of the correlation matrix R. However, this method preferably uses the row with the most non-zero elements, such as: r₂、R₁、R₀、R₁、R₂. Typically, these columns are any columns that are at least W columns from both sides, such as: at W and N_S-W-1 and any row between and including these two rows. Equation (38) and equation (39) are used in a zero-forcing equalization (zero-forcing equalization) method.

Rd^＝H^Hr equation (38)

d^＝R^-1(H^Hr) equation (39)

Since D is an orthogonal Digital Fourier Transform (DFT) matrix, equations (40), (41), and (42) can be obtained.

D^HD＝N_SI equation (40)

Equation (41)

Equation (42)

Thus, d ^ can be determined by using a Fourier Transform (FT) according to equations (43), (44), and (45 a):

equation (43)

Equation (44)

Equation (45a)

Wherein, ()₁Is the first column, although a similar equation may be applied to any column of the correlation matrix R. F () represents a Fourier transform function. F (H)^Hr) is preferably calculated using a Fast Fourier Transform (FFT) according to equation (45 b):

F(H^Hr)＝N_C[F(h₁)F(r₁)+...+F(h_N)F(r_N)]equation (45b)

Inverse Fourier transform F is performed on the result of equation (45a)^-1(.) is used to generate the spread data vector (88). In addition, the transmitted data may be recovered (90) using a despreading (despreading) method using an appropriate information code.

Table 4 shows the complexity of this Fast Fourier Transform (FFT) method.

Function of performing each burst calculation once	Number of calculations	Million instant operations/second (MROPS)
			Calculate HH		1.3224
F([R])NlogN	11600×100/10	4.4640
			Computation of H for each burst using fast Fourier transformr performs the function twice		38
Calculation equation (45)		0.8832
			F(d)NlogN		8.9280
Sum of		55MROPS

TABLE 4

This Fast Fourier Transform (FFT) method is less complex than other methods. However, this cyclic approximation may reduce performance.

For multiple chip rate sampling, another method of solving this data vector using Fast Fourier Transform (FFT) is to combine the samples using weighting, as shown in fig. 8. To facilitate explanation of the double chip rate sampling, r₁Is sampled at an even number, and r₂Is an odd number of samples. r is₁Such as: first element r₁(0) Weighted according to equation (46) and summed with r₂One corresponding element, such as: first element r₂(O), combinations.

r_eff(0)＝W₁r₁(0)+W₂r₂(0) Equation (46)

Wherein r is_effIs an effective combined matrix r_effAn effective combination of elements. W₁And W₂Is the weighting value. For N times chip rate sampling, equation (47) may be used:

r_eff(0)＝W₁r₁(0)+...+W_Nr_N(0) equation (47)

These channel response matrices H are then₁To H_NPerforms similar weighting actions to generate H_eff(92). Therefore, equation (13) can become equation (48):

r_eff＝H_effd + n equation (48)

The obtaining system is N_S×N_SA system whose system of equations can be solved (94) using Fast Fourier Transform (FFT) according to equation (49):

equation (49)

The spread data vector is determined using an inverse fourier transform. The data for the burst is determined (96) using the information codes for the bursts using a despreading method. Although equation (49) uses H_effThe first column, but the method can also be adapted to use H_effAny of which represent columns.

Another method using Fast Fourier Transform (FFT) is zero padding, which is described in detail below in conjunction with FIG. 9. According to this method, the adjustment of equation (5) fills this data vector with zero values, thereby forcing various other elements, such as: even elements, all zero (98). The adjusted matrix d is expressed as d to d. The matrix H can also be extended to the matrices H. The extension of the matrix H is repeated by each column to the right of this columnMoving each element down one row and filling the top of the shifted row with zero values. For a double chip rate sampling, for such systems, W equals 3 and N_SEqual to 4, expressed as equation (49 a):

equation (49a)

For N times chip rate sampling, equation (49b) may be used, as follows, and for ease of illustration, N_S＝3。

Equation (49b)

Generally, this N-fold matrix H-system (NN)_S)×(NN_S). This matrix H-is a square, Toeplitz, and approximate cycle, and is 2N in size_S×2N_S. This zero forcing method (ze)ro fortingsolution) is determined (100) according to equation (50):

equation (50)

Other columns than the first column may also be used for an analog Fast Fourier Transform (FFT). Furthermore, since any row can be used, the present invention can estimate a row d, which may be an estimated extended row from a row of the original channel response matrix H, or an N-fold matrix H derived from a row of the response matrix H. With the appropriate information code, d can perform despreading (desspread) to recover the data (102).

Claims (87)

1. A method for receiving a plurality of data signals transmitted from a transmitter side over a shared bandwidth in a code division multiple access communication system, each transmitted data signal experiencing a similar channel response, the method comprising:

receiving a combined signal of the transmitted data signals over the shared bandwidth;

sampling the combined signal at a multiple of a chip rate of the data signal;

estimating a channel response of the combined signal at the multiple of the chip rate;

determining a cross-correlation matrix using the estimated channel responses;

selecting a sub-block of the cross-correlation matrix;

determining a Cholesky factor for the sub-block;

extending the Cholesky factor;

determining a spread data vector using the extended Cholesky factor, a version of the channel response, and the samples; and

the data of the data signal is estimated using the spread data vector.

2. The method of claim 1 wherein the channel response is estimated as a channel response matrix and the cross-correlation matrix is a Hermetian of the channel response matrix multiplied by the channel response matrix.

3. The method of claim 2 wherein the multiple is twice the chip rate samples and the channel response matrix has even matrix samples H₁And odd matrix samples H₂。

4. The method of claim 2 wherein the multiple of the chip rate is a positive integer greater than 2.

5. The method of claim 1 wherein the sub-block has (2W-1) × (2W-1) elements of the cross-correlation matrix, and W is a length of a channel impulse response.

6. The method of claim 1 wherein the spread data vector is determined using forward and backward substitution.

7. A cdma receiver for receiving a plurality of data signals transmitted from a transmitter side over a shared bandwidth, each transmitted data signal experiencing a similar channel response, the receiver comprising:

means for receiving a combined signal of said transmitted data signals over said shared bandwidth;

means for sampling the combined signal at a multiple of a chip rate of said data signal;

means for estimating a channel response of the combined signal at the multiple of the chip rate;

means for determining a cross-correlation matrix using the estimated channel responses;

means for selecting a sub-block of the cross-correlation matrix;

means for determining a Cholesky factor for the sub-block;

means for extending the Cholesky factor;

means for determining a spread data vector using the extended Cholesky factor, a version of the channel response and the samples; and

means for estimating data of said data signal using the spread data vector.

8. The receiver of claim 7 wherein the channel response is estimated as a channel response matrix and the cross-correlation matrix is a Hermetian of the channel response matrix multiplied by the channel response matrix.

9. The receiver of claim 8 wherein the multiple is twice the chip rate samples and the channel response matrix has even matrix samples H₁And odd matrix samples H₂。

10. The receiver of claim 8 wherein the multiple of the chip rate is a positive integer greater than 2.

11. The receiver of claim 7 wherein the sub-block has (2W-1) × (2W-1) elements of the cross-correlation matrix and W is a length of the cir.

12. The receiver of claim 7 wherein the spread data vector is determined using forward and backward substitution.

13. A cdma receiver for receiving a plurality of data signals transmitted from a transmitter side over a shared bandwidth, each transmitted data signal experiencing a similar channel response, the receiver comprising:

an antenna for receiving a combined signal of said transmitted data signals over said shared bandwidth;

a sampling device for sampling the combined signal at a multiple of a chip rate of the transmitted data signal;

a channel estimation device for estimating a channel response of the combined signal at the multiple of the chip rate;

a single-user detection device that determines a cross-correlation matrix using the estimated channel response, selects a sub-block of the cross-correlation matrix, determines a Cholesky factor for the sub-block, extends the Cholesky factor, and determines a spread data vector using the extended Cholesky factor, a version of the channel response, and the samples;

wherein data of the data signal is estimated via the spread data vector.

14. The receiver of claim 13 wherein the channel response is estimated as a channel response matrix and the cross-correlation matrix is a Hermetian of the channel response matrix multiplied by the channel response matrix.

15. The receiver of claim 14 wherein the multiple is twice the chip rate samples and the channel response matrix has even matrix samples H₁And odd matrix samplingH₂。

16. The receiver of claim 14 wherein the multiple of the chip rate is a positive integer greater than 2.

17. The receiver of claim 13 wherein the sub-block has (2W-1) × (2W-1) elements of the cross-correlation matrix, and W is a length of the cir.

18. The receiver of claim 13 wherein the spread data vector is determined using forward and backward substitution.

19. A method for receiving a plurality of data signals transmitted from a transmitter side over a shared bandwidth in a code division multiple access communication system, each transmitted data signal experiencing a similar channel response, the method comprising:

receiving a combined signal of the transmitted data signals over the shared bandwidth;

sampling the combined signal at a multiple of a chip rate of the transmitted data signal;

estimating a channel response of the combined signal at the multiple of the chip rate;

determining a cross-correlation matrix using the estimated channel responses;

determining a spread data vector using sequential recursion, the spread data vector determining a first spread data vector element using an element of the cross-correlation matrix and recursively determining other spread data vector elements using other elements of the cross-correlation matrix; and

the data of the data signal is estimated using the spread data vector.

20. The method of claim 19 wherein said other spread data vector elements are determined by combining a scalar portion and a vector portion of said previously determined spread data vector elements.

21. The method of claim 19 wherein the determining the spread data vector is performed using Yule-Walker equations.

22. The method of claim 19 wherein the first spread data vector element is determined using an element in an upper left corner of the cross-correlation matrix.

23. A cdma receiver for receiving a plurality of data signals transmitted from a transmitter side over a shared bandwidth, each transmitted data signal experiencing a similar channel response, the receiver comprising:

means for receiving a combined signal of said transmitted data signals over said shared bandwidth;

means for sampling the combined signal at a multiple of a chip rate of said transmitted data signal;

means for estimating a channel response of the combined signal at the multiple of the chip rate;

means for determining a cross-correlation matrix using the estimated channel responses;

means for determining a spread data vector using sequential recursive decision, the spread data vector being determined by determining a first spread data vector element using an element of the cross-correlation matrix and recursively determining other spread data vector elements using other elements of the cross-correlation matrix; and

means for estimating data of said data signal using the spread data vector.

24. The receiver of claim 23 wherein said other spread data vector elements are determined by combining a scalar portion and a vector portion of said previously determined spread data vector elements.

25. The receiver of claim 23 wherein the determination of the spread data vector is performed using Yule-Walker equations.

26. The receiver of claim 23 wherein the first spread data vector element is determined using an element in an upper left corner of the cross-correlation matrix.

27. A cdma receiver for receiving a plurality of data signals transmitted from a transmitter side over a shared bandwidth, each transmitted data signal experiencing a similar channel response, the receiver comprising:

an antenna for receiving a combined signal of said transmitted data signals over said shared bandwidth;

a sampling device for sampling the combined signal at a multiple of a chip rate of the transmitted data signal;

a channel estimation device, said channel estimation device utilizing said multiple of said chip rate to estimate a channel response of said combined signal;

a single-user detection device that determines a cross-correlation matrix using the estimated channel responses, determines a spread data vector using sequential recursive, the single-user detection device determining a first spread data vector element using an element of the cross-correlation matrix and recursively determining other spread data vector elements using other elements of the cross-correlation matrix; and

wherein data of the data signal is estimated via the spread data vector.

28. The receiver of claim 27 wherein said other spread data vector elements are determined by combining a scalar portion and a vector portion of said previously determined spread data vector elements.

29. The receiver of claim 27 wherein the determination of the spread data vector is performed using Yule-Walker equations.

30. The receiver of claim 27 wherein the first spread data vector element is determined using an element in an upper left corner of the cross-correlation matrix.

31. A method for receiving a plurality of data signals transmitted from a transmitter side over a shared bandwidth in a code division multiple access communication system, each transmitted data signal experiencing a similar channel response, the method comprising:

receiving a combined signal of the transmitted data signals over the shared bandwidth;

sampling the combined signal at a multiple of a chip rate of the transmitted data signal;

estimating a channel response of the combined signal at the multiple of the chip rate;

determining a column of a channel correlation matrix using the estimated channel response;

determining a spread data vector using the determined column, the estimated channel response, the received combined signal, and a fourier transform; and

the data of the data signal is estimated using the spread data vector.

32. The method of claim 31 wherein the determined column is a first column of the channel correlation matrix.

33. The method of claim 31 wherein a length of an impulse response of the combined signal is W and the determined columns are at least (W-1) columns away from the boundary of the channel correlation matrix.

34. The method of claim 31 wherein the determination of the spread data vector is performed by multiplying a Hermetian matrix of a channel response matrix by a fourier transform of the received combined signal.

35. The method of claim 31 wherein the determination of the spread data vector is by a fourier transform of the determined column.

36. The method of claim 35 wherein the fourier transform of the determined column is multiplied by a number of spreading chips that are transmitted in the data signal.

37. The method of claim 31 wherein the fourier transform is a fast fourier transform.

38. The method of claim 37 wherein the determination of the spread data vector further utilizes an inverse fast fourier transform.

39. A cdma receiver for receiving a plurality of data signals transmitted from a transmitter side over a shared bandwidth, each transmitted data signal experiencing a similar channel response, the receiver comprising:

means for receiving a combined signal of said transmitted data signals over said shared bandwidth;

means for sampling the combined signal at a multiple of a chip rate of said transmitted data signal;

means for estimating a channel response of the combined signal at the multiple of the chip rate;

means for determining a column of a channel correlation matrix using the estimated channel response;

means for determining a spread data vector using the determined column, the estimated channel response, the received combined signal, and a fourier transform; and

means for estimating data of said data signal using the spread data vector.

40. The receiver of claim 39 wherein the determined column is a first column of the channel correlation matrix.

41. The receiver of claim 39 wherein a length of an impulse response of the combined signal is W and the determined columns are at least (W-1) columns away from the boundary of the channel correlation matrix.

42. The receiver of claim 39 wherein the spread data vector is determined by multiplying a Hermetian matrix of a channel response matrix by a Fourier transform of the received combined signal.

43. The receiver of claim 39 wherein the determination of the spread data vector is by a Fourier transform of the determined column.

44. The receiver of claim 43 wherein the Fourier transform of the determined column is multiplied by a number of spreading chips transmitted in the data signal.

45. A receiver according to claim 39 wherein the Fourier transform is a fast Fourier transform.

46. The receiver of claim 39 wherein the determination of the spread data vector further utilizes an inverse fast Fourier transform.

47. A cdma receiver for receiving a plurality of data signals transmitted from a transmitter side over a shared bandwidth, each transmitted data signal experiencing a similar channel response, the receiver comprising:

an antenna for receiving a combined signal of said transmitted data signals over said shared bandwidth;

a sampling device for sampling the combined signal at a multiple of a chip rate of the transmitted data signal;

a channel estimation device for estimating a channel response of the combined signal at the multiple of the chip rate; and

a single-user detection device, said single-user detection device determining a column of a channel correlation matrix using said estimated channel response, determining a spread data vector using said determined column, said estimated channel response, said received combined signal, and a fourier transform;

wherein data of the data signal is estimated via the spread data vector.

48. The receiver of claim 47 wherein the determined column is a first column of the channel correlation matrix.

49. The receiver of claim 47 wherein a length of an impulse response of the combined signal is W and the determined columns are at least (W-1) columns away from the boundary of the channel correlation matrix.

50. The receiver of claim 47 wherein the spread data vector is determined by multiplying a Hermetian matrix of a channel response matrix by a Fourier transform of the received combined signal.

51. The receiver of claim 47 wherein the determination of the spread data vector utilizes a Fourier transform of the determined column.

52. The receiver of claim 51 wherein the Fourier transform of the determined column is multiplied by a number of spreading chips transmitted in the data signal.

53. A receiver according to claim 47 wherein the Fourier transform is a fast Fourier transform.

54. The receiver of claim 47 wherein the determination of the spread data vector further utilizes an inverse fast Fourier transform.

55. A method for receiving a plurality of data signals transmitted from a transmitter side over a shared bandwidth in a code division multiple access communication system, each transmitted data signal experiencing a similar channel response, the method comprising:

receiving a combined signal of the transmitted data signals over the shared bandwidth;

sampling the combined signal at a multiple of a chip rate of the data signal;

combining the multiple chip rate samples as valid chip rate samples;

estimating a channel response of the combined signal at the multiple of the chip rate;

combining the multiple chip rate estimated channel responses as an effective chip rate channel response;

determining a spread spectrum data vector using the effective chip rate samples, the effective chip rate channel response, and a fourier transform; and

the data of the data signal is estimated using the spread data vector.

56. The method of claim 55 wherein the multiple chip rate samples and the multiple chip rate estimated channel responses are weighted prior to combining.

57. The method of claim 55 wherein the effective chip rate channel response is an effective channel response matrix.

58. The method of claim 55 wherein the determination of the spread data vector utilizes a row of a channel response matrix derived using the effective chip rate channel response.

59. The method of claim 58 wherein the row is a first row of the channel response matrix.

60. The method of claim 55 wherein the determination of the spread data vector further utilizes an inverse Fourier transform.

61. A cdma receiver for receiving a plurality of data signals transmitted from a transmitter side over a shared bandwidth, each transmitted data signal experiencing a similar channel response, the receiver comprising:

means for receiving a combined signal of said transmitted data signals over said shared bandwidth;

means for sampling the combined signal at a multiple of a chip rate of said data signal;

means for combining the multiple chip rate samples as valid chip rate samples;

means for estimating a channel response of the combined signal at the multiple of the chip rate;

means for combining the multiple chip rate estimated channel responses as an effective chip rate channel response;

means for determining a spread-spectrum data vector using the effective chip-rate samples, the effective chip-rate channel response, and a fourier transform; and

means for estimating data of said data signal using the spread data vector.

62. The receiver of claim 61 wherein the multiple chip rate samples and the multiple chip rate estimated channel responses are weighted prior to combining.

63. The receiver of claim 61 wherein the effective chip rate channel response is an effective channel response matrix.

64. The receiver of claim 61 wherein the spread data vector is determined using a row of a channel response matrix derived using the effective chip rate channel response.

65. The receiver of claim 64 wherein the row is a first row of the channel response matrix.

66. The receiver of claim 61 wherein the determination of the spread data vector further utilizes an inverse fourier transform.

67. A cdma receiver for receiving a plurality of data signals transmitted from a transmitter side over a shared bandwidth, each transmitted data signal experiencing a similar channel response, the receiver comprising:

an antenna for receiving a combined signal of said transmitted data signals over said shared bandwidth;

a sampling device for sampling the combined signal at a multiple of a chip rate of the transmitted data signal;

a channel estimation device for estimating the channel response of the combined signal at the multiple of the chip rate; and

a single-user detection device for combining the multiple chip rate samples as valid chip rate samples, for combining the multiple chip rate channel responses as a valid chip rate channel response, and for determining a spread spectrum data vector using the valid chip rate samples, the valid chip rate channel response, and a fourier transform;

wherein the data of the data signal is estimated using the spread data vector.

68. The receiver of claim 67 wherein the multiple chip rate samples and the multiple chip rate estimated channel responses are weighted prior to combining.

69. The receiver of claim 67 wherein the effective chip rate channel response is an effective channel response matrix.

70. The receiver of claim 67 wherein the spread data vector is determined using a row of a channel response matrix derived using the effective chip rate response.

71. The receiver of claim 70 wherein the row is a first row of the channel response matrix.

72. The receiver of claim 67 wherein the determination of the spread data vector further utilizes an inverse Fourier transform.

73. A method for receiving a plurality of data signals transmitted from a transmitter side over a shared bandwidth in a code division multiple access communication system, each transmitted data signal experiencing a similar channel response, the method comprising:

receiving a combined signal of the transmitted data signals over the shared bandwidth;

sampling the combined signal at a multiple of a chip rate of the data signal;

estimating a channel response using the multiple of the chip rate as a channel response matrix for the combined signal;

determining a filled version of a spread-spectrum data vector having a magnitude corresponding to the multiple chip code rate using a column of the channel response matrix, the estimated channel response matrix, the samples, and a fourier transform; and

estimating the spread data vector by removing the filled version of the elements such that the estimated spread data vector has a magnitude corresponding to the chip rate.

74. The method of claim 73 wherein the multiple of the chip rate is N, N being a positive integer greater than 1, and the estimated spread data vector includes elements of the filled version spaced apart by N elements.

75. The method of claim 74, further comprising: an extended version of the channel response matrix is determined by adding (N-1) rows to each row of the channel response matrix.

76. A method according to claim 73 wherein the Fourier transform is a fast Fourier transform.

77. The method of claim 76, wherein the determining of the filled version further utilizes an inverse fast Fourier transform.

78. A cdma receiver for receiving a plurality of data signals transmitted from a transmitter side over a shared bandwidth, each transmitted data signal experiencing a similar channel response, the receiver comprising:

means for receiving a combined signal of said transmitted data signals over said shared bandwidth;

means for sampling the combined signal at a multiple of a chip rate of said transmitted data signal;

means for estimating a channel response as a channel response matrix for the combined signal using the multiple of the chip rate;

means for determining a padded version of a spread-spectrum data vector using a column of the channel response matrix, the estimated channel response matrix, the samples, and a fourier transform, the padded version having a magnitude corresponding to the multiple chip code rate; and

means for estimating the spread data vector by removing the filled version of the elements such that the estimated spread data vector has a magnitude corresponding to the chip rate.

79. The receiver of claim 78 wherein the multiple of the chip rate is N, N being a positive integer greater than 1, and the estimated spread data vector includes elements of the filled version spaced apart by N elements.

80. The receiver of claim 79, further comprising: an extended version of the channel response matrix is determined by adding (N-1) rows to each row of the channel response matrix.

81. A receiver according to claim 78 wherein the Fourier transform is a fast Fourier transform.

82. The receiver of claim 81 wherein the padded version determination further utilizes an inverse fast Fourier transform.

83. A cdma receiver for receiving a plurality of data signals transmitted from a transmitter side over a shared bandwidth, each transmitted data signal experiencing a similar channel response, the receiver comprising:

an antenna for receiving a combined signal of said transmitted data signals over said shared bandwidth;

a sampling device for sampling the combined signal at a multiple of a chip rate of the transmitted data signal;

a channel estimation device, which estimates a channel response with the multiple of the chip rate as a channel response matrix of the combined signal; and

a single-user detection device that determines a padded version of a spread data vector having a magnitude corresponding to the multiple chip code rate using a column of the channel response matrix, the estimated channel response matrix, the samples, and a fourier transform, and estimates the spread data vector by removing elements of the padded version such that the estimated spread data vector has a magnitude corresponding to the chip code rate.

84. The receiver of claim 83 wherein the multiple of the chip rate is N, N being a positive integer greater than 1, and the estimated spread data vector includes elements of the filled version spaced apart by N elements.

85. The receiver of claim 83, further comprising: an extended version of the channel response matrix is determined by adding (N-1) rows to each row of the channel response matrix.

86. A receiver as in claim 83 wherein the Fourier transform is a fast Fourier transform.

87. The receiver of claim 86 wherein the filled version is determined by an inverse fast fourier transform.