Data detection for codes with non-uniform spreading factors

Abstract

A plurality of communication signals is received. Each communication signal has an associated code. At least two of the communication signals has a different spreading factor. The associated codes have a scrambling code period. A total system response matrix has blocks. Each block has one dimension of a length M and another dimension of a length based on in part M and the spreading factor of each communication. M is based on the scrambling code period. Data of the received plurality of communication signals is received using the constructed system response matrix.

Description

BACKGROUND

This invention generally relates to wireless code division multiple access communication systems. In particular, the invention relates to data detection of communications having non-uniform spreading factors in such systems.

In code division multiple access (CDMA) communication systems, multiple communications may be simultaneously sent over a shared frequency spectrum. Each communication is distinguished by the code used to transmit the communication. Data symbols of a communication are spread using chips of the code. The number of chips used to transmit a particular symbol is referred to as the spreading factor. To illustrate, for a spreading factor of sixteen (16), sixteen chips are used to transmit one symbol. Typical spreading factors (SF) are 16, 8, 4, 2 and 1 in TDD/CDMA communication systems.

In some CDMA communication systems to better utilize the shared spectrum, the spectrum is time divided into frames having a predetermined number of time slots, such as fifteen time slots. This type of system is referred to as a hybrid CDMA/time division multiple access (TDMA) communication system. One such system, which restricts uplink communications and downlink communications to particular time slots, is a time division duplex communication (TDD) system.

One approach to receive the multiple communications transmitted within the shared spectrum is joint detection. In joint detection, the data from the multiple communications is determined together. The joint detector uses the, known or determined, codes of the multiple communications and estimates the data of the multiple communications as soft symbols. Some typical implementations for joint detectors use zero forcing block linear equalizers (ZF-BLE) applying Cholesky or approximate Cholesky decomposition or fast Fourier transforms.

These implementations are typically designed for all the communications to have the same spreading factor. Simultaneously handling communications having differing spreading factors is a problem for such systems.

Accordingly, it is desirable to be able to handle differing spreading factors in joint detection.

SUMMARY

A plurality of communication signals is received. Each communication signal has an associated code. At least two of the communication signals has a different spreading factor. The associated codes have a scrambling code period. A total system response matrix has blocks. Each block has one dimension of a length M and another dimension of a length based on in part M and the spreading factor of each communication. M is based on the scrambling code period. Data of the received plurality of communication signals is received using the constructed system response matrix.

BRIEF DESCRIPTION OF THE DRAWING(S)

FIG. 1 is an embodiment of a non-uniform spreading factor communication system.

FIG. 2 is an illustration of a system response matrix for a k^th communication.

FIG. 3 is an illustration of constructing a total system response matrix.

FIG. 4 is a flow chart of detecting data from communications having non-uniform spreading factors.

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENT(S)

The embodiments of the invention can generally be used with any type of CDMA system, such as a TDD/CDMA, TDMA/CDMA or frequency division duplex/CDMA communication system, as well as other types of communication systems.

FIG. 1 illustrates an embodiment of a non-uniform spreading factor communication system. A transmitter 20 and a receiver 22 are shown in FIG. 1. The transmitter 20 may be located at a user equipment or multiple transmitting circuits 20 may be located at the base station. The receiver 22 may be located at either the user equipment, base station or both, although the preferred use of the receiver 22 is for use at a base station for reception of uplink communications.

Data symbols to be transmitted to the receiver 22 are processed by a modulation and spreading device 24 at the transmitter 20. The modulation and spreading device 24 spreads the data with the codes and at spreading factors assigned to the communications carrying the data. The communications are radiated by an antenna 26 or antenna array of the transmitter 20 through a wireless radio interface 28.

At the receiver 22, the communications, possibly along with other transmitters' communications, are received at an antenna 30 or antenna array of the receiver 22. The received signal is sampled by a sampling device 32, such as at the chip rate or at a multiple of the chip rate, to produce a received vector, r. The received vector is processed by a channel estimation device 36 to estimate the channel impulse responses for the received communications. The channel estimation device 36 may use a training sequence in the received communication, a pilot signal or another technique to estimate the impulse responses. A non-uniform spreading factor data detection device 34 uses the codes of the received communications and the estimated impulse responses to estimate soft symbols, d, of the spread data.

Data detection for codes having non-uniform spreading factors is illustrated in FIGS. 2 and 3 and is described with the flow chart of FIG. 4. A number, K, communications are transmitted during an observation interval. In a TDD/CDMA communication system, an observation interval is typically one data field of a communication burst. However, in TDD/CDMA as well as other CDMA communication systems, other size observation intervals may be used, such as the period of the scrambling codes.

The samples of the combined received K communications are collected over the observation interval as a received vector, r. The length in chips of r is the number of chips transmitted in the observation interval of each communication, N_c, added to the length of the channel impulse response, W, less one, (N_c+W−1).

A k^th communication of the K communications as transmitted can be represented as x^(k). An i^th chip within a symbol boundary of each symbol is defined as x_i^(k) and is per Equation 1.

$\begin{array}{cc}{\underset{\_}{x}}_i^{\left(k\right)}=\sum _{n=1}^{N_s^{\left(k\right)}}{d}_n^{\left(k\right)}{\underset{\_}{v}}_i^{\left(n,k\right)}& \mathrm{Equation}1\end{array}$

N_s^(k) is the number of symbols of the k^th communication in the observation interval. d_n^(k) is the symbol value of an n^th symbol of the N_s^(k) symbols. v^(n,k) is the portion of the code sequence of the k^th communication within the n^th symbol boundary (v^(n,k) is zero outside the n^th symbol boundary). v_i^(n,k) is the i^th chip of the portion of the code sequence within the symbol boundary (v_i^(n,k) is zero except for the i^th chip within the n^th symbol boundary).

Equation 1 can be extended to a matrix equation per Equation 2.x^(k)=V^(k)d^(k)  Equation 2

V^(k) is a spreading matrix for communication k and has N_s^(k) columns and N_c rows. An n^th column of V^(k) is v^(n,k).

After transmission through the wireless channel, x^(k) experiences a channel impulse response h^(k). h^(k) is W chips in length. h_j^(k) is a j^th chip of h^(k). Ignoring noise, the contribution, r^(k) of communication k to the received vector, r, is per Equation 3.

$\begin{array}{cc}\begin{array}{c}{\underset{\_}{r}}^{\left(k\right)}=\sum _{j=1}^W{\underset{\_}{h}}_j^{\left(k\right)}{\underset{\_}{x}}_{i-j+1}^{\left(k\right)}\\ =\sum _{j=1}^W{\underset{\_}{h}}_j^{\left(k\right)}\sum _{n=1}^{N_s^{\left(k\right)}}{\underset{\_}{d}}_n^{\left(k\right)}{\underset{\_}{v}}_{i-j+1}^{\left(n,k\right)}\\ =\sum _{n=1}^{N_s^{\left(k\right)}}{\underset{\_}{d}}_n^{\left(k\right)}\sum _{j=1}^W{\underset{\_}{h}}_j^{\left(k\right)}{\underset{\_}{v}}_{i-j+1}^{\left(n,k\right)}\end{array}& \mathrm{Equation}3\end{array}$

In matrix form, Equation 3 is per Equation 4.r^(k)=H^(k)V^(k)d^(k)  Equation 4

H^(k) is the channel response matrix for communication k and has N_c columns and (N_c+W−1) rows. The support of an i^th column of H^(k) is the channel impulse response h^(k). The first element of the support for an i^th column of H^(k) is the i^th element of that column.

For each communication k, a system transmission matrix A^(k) can be constructed per Equation 5.A^(k)=H^(k)V^(k)  Equation 5

FIG. 2 is an illustration of a system response matrix A^(k). Each column of the matrix corresponds to one data symbol of the communication. As a result, the matrix has N_s^(k) columns. Each i^th column has a block b⁽ⁱ⁾ of non-zero elements. The number of non-zero elements is determined by adding the k^th communication's spreading factor, Q_k, and the impulse response length, W, minus 1, (Q_k+W−1). The left-most column has a block b⁽¹⁾ starting at the top of the column. For each subsequent column, the block starts Q_k chips lower in the matrix. The resulting overall height of the matrix in chips is (N_c+W−1).

A total system transmission matrix can be formed by combining each communication's system response matrix A^(k), such as per Equation 6.A=[A⁽¹⁾, A⁽²⁾, . . . , A^(K)]  Equation 6

However, such a total system response matrix A would have an extremely large bandwidth. To reduce the matrix bandwidth, a block-banded toeplitz matrix is constructed, having the columns of the matrix of Equation 6 rearranged.

The height, (M+W−1), of blocks in the matrix is based on the period of the scrambling code. In many communication systems, the scrambling code repeats over a specified number of chips. To illustrate for a TDD/CDMA communication system, the scrambling code repeats after 16 chips (M=16).

A maximum spreading code of the K communications or a maximum spreading code of the communication system is referred to as Q_MAX. To illustrate, a typical TDD/CDMA communication system has a maximum spreading factor of 16 and a receiver in such a system receives communications having spreading factors of 4 and 8. In such a system, Q_MAX may be 16 (the maximum of the system) or 8 (the maximum of the received communications).

If the scrambling code period is not an integer multiple of Q_MAX, a multiple of the period may be used instead of M for constructing the blocks. To illustrate, if Q_MAX is 16 and the period is 24, three times the period (48 chips) may be used, since it is evenly divisible by 16 and 24.

Initially, columns from A⁽¹⁾, A⁽²⁾, . . . , A^(K) are selected to construct the A matrix based on each k communication's spreading factor. For the first columns of the A matrix, M/Q₁ of the first columns of A⁽¹⁾ are selected, as shown in FIG. 3. Using a second of the K matrices A⁽²⁾, M/Q₂ columns are selected. This procedure is repeated for the other K matrices, A⁽³⁾, . . . , A^(K). All of the K matrices first columns become a super column in the total system response matrix, A, having a number of columns, SC, per Equation 7, (step 100).

$\begin{array}{cc}\mathrm{SC}=\sum _{k=1}^{K}M/Q_k& \mathrm{Equation}7\end{array}$

A second super column is constructed in the same manner by selecting the next columns in the A⁽¹⁾, A⁽²⁾, . . . , A^(K) matrices. The other super columns are constructed in the same manner.

Although this illustration selects columns from the matrices in numerical order, A⁽¹⁾, A⁽²⁾, . . . , A^(K), the order of the matrices can vary. Although the resource units can be arranged in any order and still achieve a reduced bandwidth, by placing resource units transmitted with the lowest spreading factors at the exterior of each block, the bandwidth may be further reduced. However, in some implementations, the potential reduction in bandwidth may not outweigh the added complexity for reordering the K communications.

Each super column is divided into blocks having M rows, as per Equation 8, (step 102).

$\begin{array}{cc}A=\left[\begin{array}{ccccc}{B}_1& 0& \dots & 0& 0\\ ⋮& B_1& ⋰& ⋮& ⋮\\ B_L& ⋮& ⋰& 0& ⋮\\ 0& B_L& ⋰& B_1& 0\\ ⋮& 0& ⋰& ⋮& B_1\\ 0& ⋮& ⋰& B_L& ⋮\\ 0& 0& \dots & 0& B_L\end{array}\right]& \mathrm{Equation}8\end{array}$

As shown in Equation 8, the non-zero elements of each subsequent column is M rows (one block) lower in the matrix. The number of non-zero blocks in each column is L, which is per Equation 9.

$\begin{array}{cc}L=\lceil \frac{M+W-1}{M}\rceil & \mathrm{Equation}9\end{array}$ The data detection can be modeled per Equation 10.r=Ad+n  Equation 10n is the noise vector. A zero-forcing solution to Equation 10 is per Equations 11 and 12.A^Hr=Rd  Equation 11R=A^HA  Equation 12(·)^H is a complex conjugate transpose operation (Hermetian).

A minimum mean square error solution to Equation 10 is per Equations 13 and 14.A^Hr=Rd  Equation 13R=A^HA+σ²I  Equation 14is the noise variance and I is the identity matrix.

To solve either Equation 11 or 13 in a brute force approach, a matrix inversion of R, R⁻¹, is required. Using the A matrix of Equation 8, the structure of the R matrix of either Equation 12 or 14 is per Equation 15.

$\begin{array}{cc}R=\left[\begin{array}{cccccccccccc}{R}_0& R_1& R_2& R_3& R_{L-1}& 0& 0& 0& 0& 0& 0& 0\\ R_1^H& R_0& R_1& R_2& R_3& R_{L-1}& 0& 0& 0& 0& 0& 0\\ R_2^H& R_1^H& R_0& R_1& R_2& R_3& R_{L-1}& 0& 0& 0& 0& 0\\ R_3^H& R_2^H& R_1^H& R_0& R_1& R_2& R_3& R_{L-1}& 0& 0& 0& 0\\ R_{L-1}^H& R_3^H& R_2^H& R_1^H& R_0& R_1& R_2& R_3& R_{L-1}& 0& 0& 0\\ 0& R_{L-1}^H& R_3^H& R_2^H& R_1^H& R_0& R_1& R_2& R_3& R_{L-1}& 0& 0\\ 0& 0& R_{L-1}^H& R_3^H& R_2^H& R_1^H& R_0& R_1& R_2& R_3& R_{L-1}& 0\\ 0& 0& 0& R_{L-1}^H& R_3^H& R_2^H& R_1^H& R_0& R_1& R_2& R_3& R_{L-1}\\ 0& 0& 0& 0& R_{L-1}^H& R_3^H& R_2^H& R_1^H& R_0& R_1& R_2& R_3\\ 0& 0& 0& 0& 0& R_{L-1}^H& R_3^H& R_2^H& R_1^H& R_0& R_1& R_2\\ 0& 0& 0& 0& 0& 0& R_{L-1}^H& R_3^H& R_2^H& R_1^H& R_0& R_1\\ 0& 0& 0& 0& 0& 0& 0& R_{L-1}^H& R_3^H& R_2^H& R_1^H& R_0\end{array}\right]& \mathrm{Equation}15\end{array}$

As shown in Equation 15, the R matrix is block-banded and Toeplitz. As a result, solving either Equation 11 or 13 for d can be readily implemented using a block Cholesky or approximate Cholesky decomposition, (step 104). Alternately, using a circulant approximation of the R matrix of Equation 9, a block fast Fourier transform approach can be used to solve for d, (step 104).

Claims

1. A wireless communication apparatus comprising: a receiver component configured to receive and process a combined plurality of K code division multiple access (CDMA) communication signals, where each of the K communication signals has an associated code, at least two of the K communication signals have different spreading factors, and the associated codes have a scrambling code period;a data detection device configured to construct a total response matrix having a diagonal series of non-zero blocks, each block comprising a series of sub blocks that correspond to one of the K communication signals such that each block and sub block thereof have one common dimension based on the scrambling code period; andsaid data detection device further configured to use the total response matrix to estimate data of the K received plurality of communication signals.

2. The apparatus of claim 1 wherein the data detection device is configured to construct a total response matrix where said one common dimension is a multiple of the scrambling code period.

3. The apparatus of claim 1 wherein the data detection device is configured to construct a total response matrix where the diagonal series of non-zero blocks are each offset by a second common dimension.

4. The apparatus of claim 3 wherein the data detection device is configured to construct a total response matrix where the diagonal series of non-zero blocks that have said one common dimension of rows that are offset by the second common dimension of rows and where the second common dimension equals the scrambling code period.

5. The apparatus of claim 1 wherein the data detection device is configured to construct a total response matrix that defines a diagonal series of non-zero blocks that have a number of sequential columns formed from a response matrix of each respective K communication signal.

6. The apparatus of claim 1 wherein the data detection device is configured to construct a total response matrix that defines a matrix A and the data estimating means uses AHA in a zero forcing solution, where AH is a complex conjugate transpose of A.

7. The apparatus of claim 1 wherein the data detection device is configured to construct a total response matrix that defines a matrix A and the data estimating means uses AHA+σ2I in a minimum mean square error solution, where AH is a complex conjugate transpose of A, σ2 is a noise variance and I is an identity matrix.

8. The apparatus of claim 1 wherein the data detection device is configured as a base station.

9. The apparatus of claim 1 wherein the data detection device is configured as a user equipment.

10. A method for wireless communication comprising: receiving a combined plurality of K code division multiple access (CDMA) communication signals, where each of the K communication signals has an associated code, at least two of the K communication signals have different spreading factors, and the associated codes have a scrambling code period;constructing a total response matrix having a diagonal series of non-zero blocks, each block comprising a series of sub blocks that each corresponds to one of the K communication signals such that each block and sub block thereof have one common dimension based on the scrambling code period; andusing the total response matrix to estimate data of the K communication signals.

11. The method of claim 10 wherein a multiple of the scrambling code period is used as said one common dimension.

12. The method of claim 10 wherein the constructing a total response matrix defines a diagonal series of non-zero blocks that are each offset by a second common dimension.

13. The method of claim 12 wherein the constructing a total response matrix defines a diagonal series of non-zero blocks that have said one common dimension of rows that are offset the second common dimension of rows where the second common dimension equals the scrambling code period.

14. The method of claim 10 wherein the constructing a total response matrix defines a diagonal series of non-zero blocks that have a number of sequential columns formed from a response matrix of each respective K communication signal.

15. The method of claim 10 wherein the constructing a total response matrix defines a matrix A and the data estimating uses AHA in a zero forcing solution, where AH is a complex conjugate transpose of A.

16. The method of claim 10 wherein the constructing a total response matrix defines a matrix A and the data estimating uses AHA+σ2I in a minimum mean square error solution, where AH is a complex conjugate transpose of A, σ2 is a noise variance and I is an identity matrix.

17. The method of claim 10 performed by a base station.

18. The method of claim 10 performed by a user equipment.