CN100461665C

CN100461665C - Method for testing signal in multi-antenna digital wireless communication system

Info

Publication number: CN100461665C
Application number: CNB2006101040478A
Authority: CN
Inventors: 朱胡飞
Original assignee: Huawei Technologies Co Ltd
Current assignee: Huawei Technologies Co Ltd
Priority date: 2006-07-31
Filing date: 2006-07-31
Publication date: 2009-02-11
Anticipated expiration: 2026-07-31
Also published as: CN1905432A

Abstract

The invention discloses a method for detecting signals in a multi-antenna digital wireless communication system, comprising: at least two receiving antennas at a receiving end receive transmitted signal and obtain at least two received signals; the receiving end makes channel estimation according to the received signals and obtains a channel matrix H composed of channel coefficients between transmitting antenna and receiving antennas; using the channel matrix H to calculate an estimated error covariance matrix of partial transmitted signals, then using the channel matrix H and the obtained estimated error covariance matrix to recursively derive an estimated error covariance matrix of all transmitted signals; and using the derived estimated error covariance matrix of all transmitted signals to detect all the transmitted signals. And it reduces calculating complexity of signal detection.

Description

Method for detecting signal in multi-antenna digital wireless communication system

Technical Field

The present invention relates to signal detection technology, and more particularly, to a method for detecting signals in a multi-antenna digital wireless communication system.

Background

According to the information theory, the transmission bit rate can be greatly improved by simultaneously using the multi-antenna array at the transmitting end and the receiving end of the communication system.

A wireless communication system with space-time architecture using multiple antenna arrays simultaneously at the transmitting end and the receiving end is shown in fig. 1. The system operates in a rayleigh scattering environment and the individual elements of the channel matrix can be approximately considered as statistically independent. In the system shown in fig. 1, a data sequence is divided into M uncorrelated symbol subsequences, each subsequence being formed by MOne of the transmit antennas transmits. After being affected by a channel with a channel matrix of H, the M subsequences are received by N receiving antennas at a receiving end. Transmitting signal s₁，...，s_Ma-M are respectively transmitted by M different antenna units a-1₁，...，x_NReceived from N different antenna elements b-1. In the system, the number of transmitting antenna elements M is at least 2, and the number of receiving antenna elements N is at least M. The channel matrix H is an N × M matrix in which the elements in the ith row and j column represent the coupling of the ith receive antenna and the jth transmit antenna through the transmission channel. Received signal x₁，...，x_NProcessed in a digital signal processor to produce a recovered transmit signal

，...，

. Also shown in this figure are summation components c-1, c-2, c-N, which represent the contained unavoidable noise signal w₁，w₂，...，w_NThese noise signals are added to the signals received by the receiving antenna units b-1, b-2.

The matrix formed by the channel coefficients between the transmitting antennas and the receiving antennas is a channel matrix, and the channel coefficients are obtained by performing channel estimation by using the received signals. The channel matrix H in the system shown in fig. 1 is an N × M matrix, represented as:

the channel matrix H is an nxm complex matrix that is assumed to be constant over a period of K symbols. Vector h_n:(N ═ 1, 2,. cndot., N) and h_:mThe length of (M ═ 1, 2.., M) is M and N, respectively. Wherein, the channel matrix H comprises a channel vector H_:lTo h_:MRespectively representing channels for M transmission signalsThe effect of each transmitted signal in the number. More specifically, the channel vector h_:m(M ═ 1, 2.. times, M) includes channel matrix entries h_lmTo h_NmRespectively, on each of the receive antenna elements b-1 to b-N, the channel is assigned to the transmit signal s_mThe influence of (c).

In the system shown in fig. 1, the vector of the transmitted signal and the vector of the received signal satisfy the relation

Where K denotes the sampling instant, K being 1, 2, …, K. The above-mentioned relation is expressed in the form of vector

<math> <mrow> <mover> <mi>x</mi> <mo>&RightArrow;</mo> </mover> <mo>=</mo> <munderover> <mi>Σ</mi> <mrow> <mi>m</mi> <mo>=</mo> <mn>1</mn> </mrow> <mi>M</mi> </munderover> <msub> <mi>h</mi> <mrow> <mo>:</mo> <mi>m</mi> </mrow> </msub> <msub> <mi>s</mi> <mi>m</mi> </msub> <mo>+</mo> <mover> <mi>w</mi> <mo>&RightArrow;</mo> </mover> <mo>=</mo> <mi>H</mi> <mover> <mi>s</mi> <mo>&RightArrow;</mo> </mover> <mo>+</mo> <mover> <mi>w</mi> <mo>&RightArrow;</mo> </mover> <mo>,</mo> </mrow></math>

Then write the formula as

<math> <mrow> <mover> <mi>x</mi> <mo>&RightArrow;</mo> </mover> <msub> <mrow> <mo>=</mo> <mi>s</mi> </mrow> <mn>1</mn> </msub> <mo>·</mo> <msub> <mi>h</mi> <mrow> <mo>:</mo> <mn>1</mn> </mrow> </msub> <mo>+</mo> <msub> <mi>s</mi> <mn>2</mn> </msub> <mo>·</mo> <msub> <mi>h</mi> <mrow> <mo>:</mo> <mn>2</mn> </mrow> </msub> <mo>+</mo> <mo>.</mo> <mo>.</mo> <mo>.</mo> <mo>+</mo> <msub> <mi>s</mi> <mi>m</mi> </msub> <mo>·</mo> <msub> <mi>h</mi> <mrow> <mo>:</mo> <mi>m</mi> </mrow> </msub> <mo>+</mo> <mo>·</mo> <mo>·</mo> <mo>·</mo> <mo>+</mo> <msub> <mi>s</mi> <mi>M</mi> </msub> <mo>·</mo> <msub> <mi>h</mi> <mrow> <mo>:</mo> <mi>M</mi> </mrow> </msub> <mo>+</mo> <mover> <mi>w</mi> <mo>&RightArrow;</mo> </mover> </mrow></math>

In such a way that the vectors of the respective transmitted signals to the received signal are clearly seen

The influence of (c).

Minimum Mean Square Error (MMSE) estimation of the transmitted signal as

<math> <mrow> <mover> <mover> <mi>s</mi> <mo>&RightArrow;</mo> </mover> <mo>^</mo> </mover> <mo>=</mo> <msup> <mrow> <mo>(</mo> <msup> <mi>H</mi> <mi>H</mi> </msup> <mo>·</mo> <mi>H</mi> <mo>+</mo> <msub> <mi>αI</mi> <mrow> <mi>M</mi> <mo>×</mo> <mi>M</mi> </mrow> </msub> <mo>)</mo> </mrow> <mrow> <mo>-</mo> <mn>1</mn> </mrow> </msup> <msup> <mi>H</mi> <mi>H</mi> </msup> <mover> <mi>x</mi> <mo>&RightArrow;</mo> </mover> <mo>,</mo> </mrow></math>

Wherein, the symbol^-1Representing the inverse of the matrix, alpha being a constant related to the signal-to-noise ratio of the transmitted signal,

estimation error according to the invention

<math> <mrow> <mi>e</mi> <mo>=</mo> <mover> <mi>s</mi> <mo>&RightArrow;</mo> </mover> <mo>-</mo> <mover> <mover> <mi>s</mi> <mo>&RightArrow;</mo> </mover> <mo>^</mo> </mover> </mrow></math>

The covariance matrix of (1) is a covariance matrix obtained by normalizing the variance of additive white Gaussian noise to 1, i.e.

<math> <mrow> <mi>E</mi> <mrow> <mo>{</mo> <mrow> <mo>(</mo> <mover> <mi>s</mi> <mo>&RightArrow;</mo> </mover> <mo>-</mo> <mover> <mover> <mi>s</mi> <mo>&RightArrow;</mo> </mover> <mo>^</mo> </mover> <mo>)</mo> </mrow> <msup> <mrow> <mo>(</mo> <mover> <mi>s</mi> <mo>&RightArrow;</mo> </mover> <mo>-</mo> <mover> <mover> <mi>s</mi> <mo>&RightArrow;</mo> </mover> <mo>^</mo> </mover> <mo>)</mo> </mrow> <mi>H</mi> </msup> <mo>}</mo> </mrow> <mo>=</mo> <msup> <mrow> <mo>(</mo> <msup> <mi>H</mi> <mi>H</mi> </msup> <mo>·</mo> <mi>H</mi> <mo>+</mo> <msub> <mi>αI</mi> <mrow> <mi>M</mi> <mo>×</mo> <mi>M</mi> </mrow> </msub> <mo>)</mo> </mrow> <mrow> <mo>-</mo> <mn>1</mn> </mrow> </msup> <mo>,</mo> </mrow></math>

Is denoted as Q and is defined as R ═ (H)^H·H+αI_M×M) Then Q ═ R^-1. Thus, the estimate of the transmitted signal may be expressed as

<math> <mrow> <mover> <mover> <mi>s</mi> <mo>&RightArrow;</mo> </mover> <mo>^</mo> </mover> <mo>=</mo> <msup> <mi>QH</mi> <mi>H</mi> </msup> <mover> <mi>x</mi> <mo>&RightArrow;</mo> </mover> </mrow></math>

In the form of (1).

In the prior art, the method for implementing signal detection in the MIMO system described above is: firstly, an initial value of an estimation error covariance matrix Q of all to-be-detected transmission signals is obtained by using a channel matrix H, and then an estimation value of the transmission signals is calculated by using the obtained initial value of Q. How to calculate the initial value of Q will affect the amount and complexity of signal detection. The following describes a specific implementation of the prior art.

The initial value for Q is calculated in the prior art by a recursive method as follows:

first, an M diagonal matrix with (1/alpha) elements is set, namely Q_[0]＝(1/α)·I_M×MWherein Q is_[0]The first recursion variable as the initial value of recursion Q.

Then according to

Q_{[l]} = Q_{[l - 1]} - \frac{Q_{[l - 1]} h_{l :} {h_{l :}}^{H} Q_{[l - 1]}}{1 + {h_{l :}}^{H} Q_{[l - 1]} h_{l :}},

And l is 1, …, and the initial value of the estimation error covariance matrix Q corresponding to all the transmission signals to be detected is obtained by recursion of the formula of M.

The method for calculating the estimated value of the transmitting signal by using the initial value of Q comprises the following steps:

assuming that M is an iteration variable in the process of detecting the signal, the estimation error covariance matrix corresponding to the M +1-M to-be-detected transmission signals when the mth transmission signal is detected is recorded as Q_M+1-m。

Q when m is 1, i.e. when the first transmitted signal is detected_M+1-mIs the initial value of Q, so the estimated value of the first detected transmission signal is calculated directly using the initial value of Q. Then, obtaining an estimation error covariance matrix Q corresponding to the remaining M-1 to-be-detected transmission signals by utilizing Q recursion_M-1。

When m is>1, the estimation error covariance matrix Q corresponding to M +1-M to-be-detected transmission signals_M+1-mCalculated after last detection of the transmitted signal and using Q_M+1-mAnd calculating to obtain an estimated value of the m-th detected transmitting signal. Then, using Q_M+1-mRecursion is carried out to obtain an estimation error covariance matrix Q corresponding to the rest M-M to-be-detected transmission signals_M-m。

In summary, the matrix in the recursive process of the initial value of Q required in signal detection is an M × M matrix from the beginning, and since the computational complexity of the matrix is directly related to the matrix dimension, the recursive computational complexity becomes higher when the number of transmit antennas M is large. In addition, in the signal detection process, the estimation error covariance matrix corresponding to the current transmission signal to be detected needs to be recalculated every time, which inevitably leads to the increase of the signal detection calculation complexity.

Disclosure of Invention

It is therefore an objective of the claimed invention to provide a method for detecting signals in a multi-antenna digital wireless communication system, so that the computational complexity of detecting signals is reduced.

In order to achieve the above object, the present invention provides a method for detecting signals in a multi-antenna digital wireless communication system, which detects at least two transmission signals in a MIMO system, wherein the transmission signals are transmitted from different transmission antennas of a transmitting end and reach a receiving end through a channel, the method comprising: a. at least two receiving antennas of the receiving end receive the transmitting signals to obtain at least two receiving signals; b. the receiving end carries out channel estimation according to the received signal to obtain a channel matrix H consisting of channel coefficients between the transmitting antenna and the receiving antenna; c. calculating an estimation error covariance matrix of partial transmitting signals in all transmitting signals by using the channel matrix H, and then recursively obtaining the estimation error covariance matrix of the transmitting signals which comprise the partial transmitting signals and the number of which is more than the number of the partial transmitting signals by using the channel matrix H and the calculated estimation error covariance matrix of the partial transmitting signals; wherein the number of rows and columns of the estimation error covariance matrix of the partial transmit signals is equal to the number of the partial transmit signals, and the number of rows and columns of the estimation error covariance matrix of the transmit signals including the partial transmit signals and having a number greater than the number of the partial transmit signals is equal to the number of the transmit signals including the partial transmit signals and having a number greater than the number of the partial transmit signals; d. and c, detecting the transmitting signals which comprise the partial transmitting signals and the number of which is more than that of the partial transmitting signals in the step c by using the estimation error covariance matrix obtained in the step c.

C, said recursion obtaining the estimation error covariance matrix of the transmitting signals which comprise the partial transmitting signals and the number of the transmitting signals is more than the number of the partial transmitting signals is: recursion is carried out to obtain the covariance matrix of the estimation errors of all the transmitted signals; step d the detection comprises partial emission signals and the number of the emission signals which is more than the number of the partial emission signals is: all transmitted signals are detected.

Step c, the step of obtaining the estimated error covariance matrixes of all the transmitting signals by recursion of the estimated error covariance matrixes of part of the transmitting signals comprises the following steps: and recursion is carried out to obtain the estimation error covariance matrixes of all the transmitted signals by taking Sherman-Morrison results of the estimation error covariance matrixes of part of the transmitted signals as sub-matrixes.

The step b and the step c further comprise the following steps: setting and detecting the sequence of at least two transmitting signals;

the step c comprises the following steps:

c11. calculating a covariance matrix of estimation errors of the first number of transmission signals detected at last by using a channel matrix corresponding to the first number of transmission signals detected at last in the set detection sequence;

setting a second number for recursively detecting the transmit signal estimation error covariance matrices greater than the first number;

c12. and c, recursion calculating the estimated error covariance matrix of the second number of finally detected transmitting signals by utilizing the channel matrix corresponding to the second number of finally detected transmitting signals in the set detection sequence and taking the Sherman-Morrison result of the estimated error covariance matrix of the first number of finally detected transmitting signals obtained in the last recursion or step c11 as a sub-matrix, if the estimated error covariance matrices of all the detected transmitting signals are obtained, ending the process, otherwise, adding an integer value of 1 or more than 1 to the value of the second number after the value of the first number is equal to the value of the second number, and returning to the step c12.

The step d comprises the following steps:

d1. c, selecting a currently detected transmitting signal from the transmitting signals to be detected, and obtaining an estimated value of the currently detected transmitting signal by using the estimated error covariance matrix, the channel matrix H and the received signals of all the transmitting signals obtained in the step c;

d2. calculating an interference value for detecting the subsequent emission signal to be detected by using the estimated value of the currently detected emission signal obtained in the step d1, and eliminating the interference of the currently detected emission signal on the detection of the subsequent emission signal to be detected;

d3. repeating the steps d1, d2 until all emission signals to be detected are detected.

The step d1 further comprises: carrying out pre-matched filtering transformation on the received signal by utilizing a channel matrix H; computing the cross-correlation of the channel matrix HChannel matrix phi, phi ═ H^H·H；

Step d1 wherein the step of obtaining an estimate of a currently detected transmitted signal comprises: obtaining an estimated value of the currently detected transmitting signal by utilizing an estimated error covariance matrix of the transmitting signal to be detected and a pre-matched filtering result of the receiving signal;

the step d2 includes: and calculating the interference value of the detected transmitting signal to the detection of the subsequent transmitting signal by utilizing the estimated value of the currently detected transmitting signal and the cross-correlation channel matrix phi of the channel matrix H, and eliminating the interference of the detected transmitting signal from the pre-matched filtering result of the receiving signal to obtain the corrected pre-matched filtering result of the receiving signal, wherein the corrected pre-matched filtering result is used as the pre-matched filtering result of the receiving signal when the signal is detected next time.

According to the signal detection method provided by the invention, when the initial values of the estimated error covariance matrixes of all the transmitting signals are calculated, the sequence of detecting the transmitting signals at the receiving end is preset, and the estimated error covariance matrixes corresponding to the transmitting signals from less to more are recurred in sequence according to the sequence, so that the initial values of the estimated error covariance matrixes of all the transmitting signals are finally obtained. The intermediate result obtained in the recursion process of the initial value can also be utilized in the signal detection process, if the preset detection sequence is similar to or the same as the actual detection sequence, the step of recalculating the estimation error covariance matrix corresponding to the to-be-detected transmitting signal is reduced in the signal detection process, and therefore the calculation complexity of the detection signal can be reduced.

Drawings

FIG. 1 is a block diagram of a prior art multi-antenna digital wireless communication system;

FIG. 2 is a flow chart of the present invention for calculating the initial values of the estimated error covariance matrices for all transmitted signals;

fig. 3 is a flow chart showing the detection of signals in the present invention.

Detailed Description

In order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention is described in further detail below with reference to specific embodiments.

In the process of detecting signals, the main idea of calculating the initial values of the estimation error covariance matrixes of all the transmitted signals is as follows: according to the sequence of the less transmitted signals, firstly, the estimated error covariance matrix of the less transmitted signals is obtained through calculation, and then the estimated error covariance matrix of the more transmitted signals is calculated through recursion by utilizing the obtained estimated error covariance matrix of the less transmitted signals. By calculating the initial value of the estimation error covariance matrix by the above recursive method, not only the calculation amount of the recursive initial value can be reduced, but also the intermediate result obtained in the process of calculating the initial value can be fully utilized in the step of detecting the signal, thereby further reducing the calculation amount of the detection signal.

Specific implementation methods are given below. Fig. 2 is a flow chart showing the calculation of initial values of the estimation error covariance matrices for all transmitted signals, comprising the following steps:

step 201: the receiving end receives M transmitting signals respectively transmitted by the transmitting end from M transmitting antennas to obtain N receiving signals, and performs channel estimation according to the receiving signals to obtain a channel matrix H consisting of channel coefficients between the transmitting antennas and the receiving antennas.

Presetting the sequence of all M transmitting signals detected at the receiving end, and marking the sequence number of the transmitting signals as t_M，t_M-1，…，t_m，…，t₂，t₁. Correspondingly, the channel matrix H is reordered according to columns to obtainThe preset channel matrix H for detecting the sequencing of the transmitted signals is recorded as

Using vector f ═ t₁，t₂，…，t_m，…，t_M-1，t_M]^TRecording and channel matrix

The index of the corresponding transmitted signal.

Step 202: by means of channel matrices

First, obtainCross correlation channel matrix of

Then from phi_MSolving a covariance matrix Q of the estimation error_MInverse matrix of

Wherein,

wherein,^*indicating the conjugation of 1 complex number.

Step 203: calculating a last detected transmission signal t₁Corresponding estimation error covariance matrix, noted

Corresponding to the transmitted signal t₁The channel matrix of

H_{1}^{(t_{1})} = [h_{{: t}_{1}}] .

R from step 202_MIn (1), obtaining a transmission signal t₁The inverse of the estimated error covariance matrix of

Wherein,

is R_MRow 1, column 1 elements.

Then use

Calculating the corresponding last detected transmission signal t₁Of the estimated error ofBy

Q_{1}^{(t_{1})} = {(R_{1}^{(t_{1})})}^{- 1}

To obtain

Q_{1}^{(t_{1})} = 1 / r_{t_{1} t_{1}} .

Next, the variable m of the initial value of the recurrence Q is set to 2, and the process proceeds to step 204. In the following step of recursion of the initial value of Q, the m transmission signals t to be detected last_m，…，t₂，t₁The corresponding estimation error covariance matrix is noted as

Step 204: judging whether the estimated error covariance matrixes of all the detected transmitting signals are obtained or not, namely judging whether M is larger than M or not, if so, indicating that the estimated error covariance matrixes of M detected transmitting signals are obtained, and turning to the step 208; otherwise, recursion of the estimated error covariance matrix of m detected

transmission signals

Step

205, 206, 207 are performed.

Step 205: m transmission signals t detected at last_m，…，t₂，t₁The corresponding channel matrix is

Correspondingly, corresponding to the m transmission signals t detected last_m，…，t₂，t₁The inverse of the estimated error covariance matrix of

And

the following recurrence relation:

wherein,

is the last recursion to obtain m-1 transmission signals t corresponding to the last detection_m-1，…，t₂，t₁Or the inverse of the estimated error covariance matrix of (1) or the 1 transmission signal t corresponding to the last detection₁Is the inverse of the covariance matrix of the estimated error

Is easy to see

And

r can be calculated from step 202_MIs directly obtained, more specifically is R_MThe m-th row and the m-th column of

Is formed by R_MThe first m-1 item of the m-th column of (1). Thereby directly obtaining the result without any calculation

Step 206: finding the m finally detected transmission signals t_m，…，t₂，t₁Corresponding estimation error covariance matrix

Using estimated error covariance matrix corresponding to last detected transmitted signalOr the last recursion to obtain m-1 transmitting signals t corresponding to the last detected transmitting signals_m-1，…，t₂，t₁Of the estimated error covariance matrix

Recursion to obtain

The recursion method is as follows:

first of all, calculate

The Sherman-Morrison result of (A) is obtained by using the Sherman-Morrison formula

Then is made of

And

m transmitting signals t detected finally are obtained_m，…，t₂，t₁Corresponding estimation error covariance matrix

In order to realize the purpose,

step 207: the value of m is increased by 1, i.e., m +1, and then the process returns to step 204.

Step 208: obtaining all M transmission signals t_M，t_M-1，…，t_m，…，t₂，t₁Corresponding estimation error covariance matrix

The value of (c).

That is, an optimal detection sequence is determined in the signal detection process, and when each transmission signal is detected successively according to the optimal detection sequence and using the interference cancellation method, the initial value of the estimation error covariance matrix is used, and the initial value is denoted as QM,

Q_{M} = Q_{M}^{(t_{M})} .

through the steps, the initial value Q of the estimation error covariance matrix of all the detected transmitting signals is obtained_M(ii) a Meanwhile, an estimation error covariance matrix Q of the emission signal of the detection part is obtained_mWherein, M is 1, 2_mCorresponding to the transmitting antenna t_m，…，t₂，t₁The estimated error covariance matrix of (2).

After the initial values of the estimation error covariance matrices of all the transmission signals to be detected are obtained, the process of detecting the signals shown in fig. 3 is entered, that is, turn to a in fig. 3.

Fig. 3 is a flow chart of signal detection, and the signal detection shown in fig. 3 starts from a.

Step 301: carrying out pre-matched filtering transformation on the received signal to obtain the pre-matched filtering result of the received signal

<math> <mrow> <msub> <mi>z</mi> <mi>M</mi> </msub> <mo>=</mo> <msup> <mrow> <mo>(</mo> <msub> <mi>H</mi> <mi>M</mi> </msub> <mo>)</mo> </mrow> <mi>H</mi> </msup> <mo>·</mo> <mover> <mi>x</mi> <mo>&RightArrow;</mo> </mover> <mo>,</mo> </mrow></math>

Where (HM) H is a matched filter, vector

To represent a received signal x₁，...，x_NThe vector of (2). The index of the transmitted signal is still the vector f ═ t₁，t₂，…，t_m，…，t_M-1，t_M]^T。

Next, an iteration variable m in the process of detecting a signal is set to 1, and the process proceeds to step 302. In the following iteration step of detecting signals, when a certain transmitting signal is detected in M +1-M transmitting signals, the covariance matrix of the estimation errors corresponding to the M +1-M transmitting signals to be detected is recorded as Q_M+1-m。

Step 302: determining whether the last transmitted signal was detected, i.e., determining whether M equals M, and if so, performing step 313; otherwise step 303 is performed.

Step 303: determining the transmitting signal with the best receiving signal-to-noise ratio in the M +1-M transmitting signals, wherein the method comprises the following steps: from Q_M+1-mIn search pairThe row with the smallest corner element is denoted as the l_mThe rows of the image data are, in turn,

l_{m} = \underset{i}{\arg \min} q_{M + 1 - m, ii},

wherein q is_M+1-m，iiIs a matrix Q_M+1-mI.e., the element on the diagonal line, i.e., 1, i.m + 1-M. The first mentioned_mThe row corresponds to the signal with the best received signal-to-noise ratio of the M +1-M transmitted signals, i.e. the currently detected transmitted signal.

Step 304: in the matrix Q_M+1-mMiddle exchange of_mLines and M +1-M lines, exchange the l_mColumns and M +1-M columns; correspondingly, in the matrix R_M+1-mMiddle exchange of_mLines and M +1-M lines, exchange the l_mColumns and M +1-M columns; correspondingly, in the matrix phi_M+1-mMiddle exchange of_mLines and M +1-M lines, exchange the l_mColumns and M +1-M columns; accordingly, in the vector z of the result of the pre-matched filtering of the received signal_M+1-mMiddle exchange of_mItems and M +1-M items. Exchanging the l-th in the vector f_mItems and M +1-M items.

Step 305: the currently detected emission signal is the M +1-M term in the vector f and is marked as p_m，p_mF (M + 1-M). Calculating an estimate of the currently detected transmitted signal

Transmitting signal p_mIs estimated value of

In order to realize the purpose,

wherein q is_M+1-mRepresents Q_M+1-mColumn M + 1-M.

Step 306: estimation of the resulting transmitted signal

Quantizing to obtain the detection result of the transmitted signal

Step 307: eliminating the influence of the currently detected transmitting signal from the pre-matched filtering result of the receiving signal to obtain the pre-matched filtering results of a plurality of receiving signals corresponding to M-M transmitting signals which are not detected, namely: from vector z of results of pre-matched filtering of received signal_M+1-mWherein deletion of the M +1-M entry results in (z) having M-M entries_M+1-m)^minus(ii) a From (z)_M+1-m)^minusThe interference of the currently detected transmitting signal is eliminated to obtain the result z of the pre-matched filtering of a plurality of receiving signals corresponding to all M-M undetected transmitting signals_M-m，Wherein

Is a matrix phi_M-m+1The first M-M rows of the M +1-M columns,

having an M-M term.

Step 308: from matrix Φ_M+1-mDeleting the M +1-M column and the M +1-M row to obtain a matrix phi_M-m(ii) a From matrix R_M+1-mExtracting M-M row heads and M-M columns of the middle extraction head to obtain a matrix R_M-mFrom the matrix R_M+1-mExtracting the first M-M rows of the M +1-M columns to obtain a vector v_M-mFrom the matrix R_M+1-mExtracting the elements in the M +1-M row and the M +1-M column to obtain an item

Step 309: the judgment is made to find Q as shown in FIG. 2_MWhether an estimated error covariance matrix Q corresponding to M-M transmission signals has been obtained or not in the process of the initial value of (a)_M-mIf yes, go to step 310; otherwise, step 311 is performed.

Step 310: will recur Q_MQ obtained in the course of the initial value of (1)_M-mAs an estimated error covariance matrix Q for the M-M transmit signals required for the next iteration_M-m. Step 312 is then performed.

Step 311: using a matrix Q_M+1-mCalculating an estimated error covariance matrix Q corresponding to the M-M transmitted signals required for the next iteration_M-mThe calculation method comprises the following steps: slave matrix Q_M+1-mDeleting the M +1-M row and the M +1-M column to obtain a matrixThen use

v_M-mAnd

to obtain Q_M-m，

Step 312 is then performed.

Step 312: the value of m is increased by 1, i.e., m is m +1, and the process returns to step 302.

Step 313: the currently detected transmission signal is the last term of the vector f, denoted as p_M. Calculating an estimate of the last detected transmitted signal

Transmitting signal p_MIs estimated value of

In order to realize the purpose,

wherein q is₁Is namely Q₁。

Step 314: estimation of the resulting transmitted signal

Quantizing to obtain the detection result of the transmitted signal

By the above method, the detection order of the transmission signals is p₁，p₂，…，p_MCorrespondingly, the detection result of the transmitted signal is

，

，…，

。

In the above embodiment, it can be seen that, if the preset sequence of the detection transmit signals when the initial value of the error covariance matrix Q is estimated by recursion of the transmit signals is the same as the sequence of the actual detection transmit signals, the method described in step 311 is not used to calculate and obtain the estimated error covariance matrix required by the next iteration, but the estimated error covariance matrix required by the next iteration is directly obtained from the intermediate result obtained in the process of recursion of the initial value of Q, so that a lot of calculation amount can be reduced; alternatively, if the preset sequence of the detection transmitting signals when the initial value of Q is recurred is close to the sequence of the actual detection transmitting signals, the step 311 can be omitted, so that the amount of calculation can be reduced.

Therefore, in the slow fading channel, the sequence of detecting M transmitting antennas by the receiving end may be set as the optimal sequence of the last detection, so that in the process of detecting signals shown in fig. 3, the amount of calculation in step 311 is reduced. In a slow fading channel, the channel characteristics change slowly, and compared with the most recent optimal detection sequence, the optimal detection sequence at the current moment does not change much or the same, so that the calculation amount of the detection signal can be reduced by well utilizing an intermediate result obtained in the process of recursion of the initial value of Q.

In fast fading channels, there are also many existing techniques, and a detection order can be estimated by the channel matrix H, so that the detection order is close to the optimal detection order actually used.

In some applications, the order of detecting all the transmitted signals is fixed in advance, and the transmitted signals are detected one by one according to the fixed detection order, so that the optimal detection order is not required in the process. In this case, the sequence in which all M transmission signals preset in step 201 are detected at the receiving end is the predetermined detection sequence, so that the process of determining the detection sequence as described in step 303 is not required, and which signal to be detected is currently selected for detection is determined according to the predetermined detection sequence; thus, no exchange of rows and columns of the matrix as described in step 304 is required; meanwhile, the process of calculating the estimation error covariance matrix required for the next iteration as described in step 311 is not needed, and the intermediate result obtained in the recursive Q process is directly used to detect the next to-be-detected transmission signal. By the above method, the amount of calculation can be reduced much.

The above description is only for the purpose of illustrating the preferred embodiments of the present invention and is not to be construed as limiting the invention, and any modifications, equivalents, improvements and the like that fall within the spirit and principle of the present invention are intended to be included therein.

Claims

1. A method for detecting signals in a multi-antenna digital radio communication system, in which at least two transmit signals are detected in a multiple-input multiple-output, MIMO, system, said transmit signals being transmitted by different transmit antennas at a transmit end and via a channel to a receive end, characterized in that,

a. at least two receiving antennas of the receiving end receive the transmitting signals to obtain at least two receiving signals;

b. the receiving end carries out channel estimation according to the received signal to obtain a channel matrix H consisting of channel coefficients between the transmitting antenna and the receiving antenna;

c. calculating an estimation error covariance matrix of partial transmitting signals in all transmitting signals by using the channel matrix H, and then recursively obtaining the estimation error covariance matrix of the transmitting signals which comprise the partial transmitting signals and the number of which is more than the number of the partial transmitting signals by using the channel matrix H and the calculated estimation error covariance matrix of the partial transmitting signals; wherein the number of rows and columns of the estimation error covariance matrix of the partial transmit signals is equal to the number of the partial transmit signals, and the number of rows and columns of the estimation error covariance matrix of the transmit signals including the partial transmit signals and having a number greater than the number of the partial transmit signals is equal to the number of the transmit signals including the partial transmit signals and having a number greater than the number of the partial transmit signals;

d. and c, detecting the transmitting signals which comprise the partial transmitting signals and the number of which is more than that of the partial transmitting signals in the step c by using the estimation error covariance matrix obtained in the step c.

2. The method of claim 1, wherein said step c of recursively calculating estimated error covariance matrices for more than one transmit signal including said partial transmit signal is: recursion is carried out to obtain the covariance matrix of the estimation errors of all the transmitted signals; step d the detection comprises partial emission signals and the number of the emission signals which is more than the number of the partial emission signals is: all transmitted signals are detected.

3. The method of claim 2, wherein the step c of recursively obtaining the covariance matrix of the estimated errors of all the transmitted signals by using the covariance matrix of the estimated errors of the partial transmitted signals comprises: and recursion is carried out to obtain the estimation error covariance matrixes of all the transmitted signals by taking Sherman-Morrison results of the estimation error covariance matrixes of part of the transmitted signals as sub-matrixes.

4. The method of claim 2, further comprising between step b and step c: setting and detecting the sequence of at least two transmitting signals;

the step c comprises the following steps:

5. The method of claim 2,

the step c comprises the following steps:

c21. calculating a covariance matrix of an estimation error of the last detected transmitting signal by using a channel matrix corresponding to the last detected transmitting signal in the set detection sequence;

c22. using a channel matrix corresponding to the m transmitting signals detected last in the set detection sequence, and recurrently calculating the estimated error covariance matrixes of the m transmitting signals detected last by taking a Sherman-Morrison result of the estimated error covariance matrixes of the m-1 transmitting signals detected last obtained in the step c21 or recurrently calculating the estimated error covariance matrixes of the m transmitting signals detected last as a sub-matrix, if the estimated error covariance matrixes of all the transmitting signals are obtained, ending the step, otherwise adding 1 to the value of m, and returning to the step c 22;

where m is set to an initial value of 2.

6. The method of claim 5,

the number of the transmitting antennas is M, the number of the receiving antennas is N, and M different transmitting antennas respectively transmit M transmitting signals;

the step of setting and detecting the sequence of at least two emission signals comprises the following steps: reordering the M transmitted signals transmitted by the transmitting antennas to obtain said order, denoted t by the sequence of the transmitted signals_M，t_M-1，…，t_m，…，t₂，t₁；

The step c21 includes: using a transmission signal t₁Corresponding channel matrixObtaining an inverse of an estimated error covariance matrix of the transmitted signal

And according to the transmission signal t₁And the estimated error covariance matrix of

The relation satisfied by the matrix obtains the transmitting signal t₁Estimating an error covariance matrix of (a);

the step c22 includes: using m transmitted signals t₁…t_mCorresponding channel matrix

Obtaining an inverse matrix of the estimated error covariance matrices of the m transmit signals

Is not included in

Is to be used for the partial item in (b),and according to the estimated error covariance matrix of the m transmitting signals

Satisfied relation, and use of the resulting m-1 transmission signals t₁…t_m-1Recursion of the estimated error covariance matrixes of the M transmitted signals, and ending the step if the estimated error covariance matrixes of the M detected signals are obtained; otherwise, adding 1 to the value of m, and returning to execute the step c 22;

wherein,

representing the sum of the transmitted signal t in the channel matrix H_iThe corresponding column vector, i ═ 1 … M.

7. The method of claim 6,

step c21

Wherein α is a constant related to the signal-to-noise ratio of the transmitted signal;

step c22Is not included in

The partial items in (1) are: a scalar quantity

And a vectorWherein,

the estimated error covariance matrices of the m transmitted signals obtained by the recursion in step c22 are: and adding a matrix obtained by one row and one column on the basis of Sherman-Morrison results of the covariance matrix of the estimation errors of the m-1 transmitting signals.

8. The method of claim 7,

step c22, on the basis of the Sherman-Morrison result of the covariance matrix of the estimated errors of m-1 transmitted signals, adding a row and a column to obtain a matrix: covariance matrix of estimated errors of m-1 transmitted signalsThe Sherman-Morrison result is

In that

On the basis of the basic matrix, the channel matrix is added

(Vector)

And a scalar quantity

Forming a column and a row to obtain an estimated error covariance matrix of m transmitted signals

Wherein one column and one row of intersecting terms is

Other items of a column are

Other terms of a row are

Wherein,for the result of the last recursion or obtained in step c21

9. The method of claim 2, wherein step d comprises:

10. The method of claim 9,

the step d1 further comprises: carrying out pre-matched filtering transformation on the received signal by utilizing a channel matrix H; calculating the cross-correlation channel matrix phi of the channel matrix H, phi ═ H^H·H；

11. The method of claim 10,

the step of calculating the cross-correlation channel matrix Φ of the channel matrix H comprises: calculating an inverse matrix R of an estimated error covariance matrix of the transmitted signal using a channel matrix H, using phi ═ H^HH and R ═ H^H·H+αI_M×MThe relationship of (3) gives phi.

12. The method of claim 10,

the step of performing the pre-matched filtering transformation on the received signal by using the channel matrix H comprises the following steps: taking the conjugate transpose matrix of the channel matrix H as a pre-matched filter of the received signal, and performing pre-matched filtering on the vector of the received signal to obtain a pre-matched filtering result of the received signal;

step d1, selecting the currently detected one of the emission signals to be detected as: the transmitting signal corresponding to the row with the minimum diagonal element in the estimation error covariance matrix of the transmitting signal to be detected is a currently detected transmitting signal;

the step d1 of obtaining the estimated value of the currently detected transmitting signal by using the covariance matrix of the estimated error of the transmitting signal to be detected and the result of the pre-matched filtering of the receiving signal includes: multiplying the row with the minimum diagonal element in the estimation error covariance matrix of the transmitting signal to be detected by the pre-matched filtering result of the receiving signal to obtain the estimation value of the currently detected transmitting signal;

the step d2 includes: and obtaining an interference value of the detected transmitting signal to the detection of the subsequent transmitting signal according to the product of the estimated value of the currently detected transmitting signal and a vector formed by elements corresponding to the currently detected transmitting signal in the cross-correlation channel matrix phi of the channel matrix, deleting an item corresponding to the detected transmitting signal from the pre-matched filtering result of the receiving signal, and then eliminating the interference from the pre-matched filtering result of the received signal after deleting the item to obtain a corrected pre-matched filtering result of the receiving signal.

13. The method of claim 12, wherein after step d2, before detecting the next transmitted signal, the method further comprises: judging whether an estimated error covariance matrix corresponding to the to-be-detected transmission signal is obtained in the process of recursion of the estimated error covariance matrices of all the transmission signals in the step c, if so, detecting the next transmission signal by using the estimated error covariance matrix corresponding to the to-be-detected transmission signal in the next detection signal obtained in the process of recursion of the estimated error covariance matrices of all the transmission signals in the step c; otherwise, the estimation error covariance matrix corresponding to the to-be-detected transmitting signal in the current signal detection is used for recursion to obtain the estimation error covariance matrix corresponding to the to-be-detected transmitting signal in the next signal detection.

14. The method according to claim 13, wherein the estimating error covariance matrix corresponding to the to-be-detected transmission signal in the next signal detection is obtained by recursion using the estimation error covariance matrix corresponding to the to-be-detected transmission signal in the current signal detection: and deleting the Sherman-Morrison result of the submatrix obtained by the row and the column with the minimum diagonal elements from the estimation error covariance matrix corresponding to the to-be-detected emission signal during the current signal detection, and taking the Sherman-Morrison result as the estimation error covariance matrix corresponding to the to-be-detected emission signal during the next signal detection.

15. The method according to claim 1, wherein in step d, when detecting one of the transmission signals to be detected, it is determined whether the estimated error covariance matrix corresponding to the transmission signal to be detected is obtained in step c during the process of obtaining the estimated error covariance matrix of the transmission signals including the partial transmission signals and having the number greater than the number of the partial transmission signals, if so, the transmission signal is detected by using the estimated error covariance matrix corresponding to the transmission signal to be detected obtained in step c during the process of obtaining the estimated error covariance matrix of the transmission signals including the partial transmission signals and having the number greater than the number of the partial transmission signals,

the number of the current emission signals to be detected is less than the number of the emission signals which comprise the partial emission signals and the number of the emission signals is more than the number of the partial emission signals.