阅读说明：本技术 一种基于对称lq的大规模mimo系统迭代信号检测方法 (Large-scale MIMO system iterative signal detection method based on symmetry L Q ) 是由 景小荣 陈洪燕 于 2020-04-14 设计创作，主要内容包括：本发明涉及一种基于对称LQ的大规模MIMO系统迭代信号检测方法,属于无线通信技术领域。该方法首先将大规模MIMO系统中多用户信号恢复问题转换为求解线性方程组；然后采用对称LQ(Symmetric LQ,S-LQ)方法来迭代求解线性方程组,完成最大迭代次数后,将线性方程组的解向量作为发送信号矢量的估计值。与传统MMSE线性检测算法相比,本发明规避了计算MMSE滤波所需的高维矩阵求逆,从而大幅度降低了计算复杂度,同时取得了接近MMSE检测算法的性能。(The invention relates to a large-scale MIMO system iterative signal detection method based on symmetry L Q, belonging to the technical field of wireless communication, the method firstly converts a multi-user signal recovery problem in a large-scale MIMO system into a linear equation solving system, then adopts a symmetry L Q (symmetry L Q, S-L Q) method to iteratively solve the linear equation system, and takes a solution vector of the linear equation system as an estimated value of a transmitted signal vector after the maximum iteration times are completed.)

1. A large-scale MIMO system iterative signal detection method based on symmetry L Q is characterized by comprising the following steps:

s1: converting a multi-user signal recovery problem in a large-scale MIMO system into a linear equation solving system;

and S2, iteratively solving the linear equation set by adopting a Symmetric L Q (symmetry L Q, S-L Q) method, and taking a solution vector of the linear equation set as an estimated value of a transmission signal vector after the maximum iteration times are finished.

2. The iterative signal detection method for the symmetric L Q-based massive MIMO system according to claim 1, wherein the step S1 specifically comprises assuming that a base station is equipped with N receiving antennas to serve K single-antenna users in the uplink multiuser massive MIMO system, receiving a signal vector y ═ Hx + N by the base station, receiving with Minimum Mean-Square Error (MMSE), and transmitting an estimated value of the vector x by the userWherein F ═ H^HH+σ²I_K)^-1H^HAn MMSE equalization matrix; g is H^HH，W＝G+σ²I_K，Then satisfyThen, the problem of signal detection is converted into the solution of a linear equation set; wherein y is_N×1Is receivingSignal vector, x_K×1Is a transmitted signal vector, H_N×KIs a channel matrix; n is a noise vector, assumed to obey a mean of 0 and a covariance matrix of σ²I_NComplex Gaussian random variable of (I)_NIs an N × N-dimensional unit matrix;is an estimate of the transmitted signal vector, W is the MMSE filtering matrix,is a matched filtered signal; (.)^HIndicating the operation of conjugate transpose of the matrix, superscript (. cndot.)^-1Representing the inversion of the matrix.

3. The iterative signal detection method for massive MIMO system based on symmetry L Q of claim 1, wherein the step S2 comprises applying a symmetry L Q method to solve the linear system of equations iterativelyIn the t +1 th iteration, the solution vector is updated:

wherein, g^(t+1)、c^(t+1)And ζ^(t+1)As an iteration coefficient, w^(t)And v^(t+1)For iterative vector, superscript (. cndot.)^(t)Representing the T iteration, T ∈ 1,2, theSolution vector ofAs an estimate of the transmitted signal vector.

4. The symmetrical L Q-based large-scale MIMO system iterative signal detection method of claim 3, wherein a symmetrical L Q method is adopted to solve the linear equation system iterativelyThe specific process comprises the following steps:

let w and v be two sets of linearly independent vectors, let K_m＝span w，K_mIs the right subspace, L_m＝span v，L_mIs the left subspace; for linear system of equationsIterative solution is carried out to seek a value belonging to K_mApproximate solution ofThe condition of Petorv-Galerkin is satisfied:

by using the dominant diagonal dominance characteristic of the matrix W, the initial solution of the S-L Q iterative method is set asThe process of solving the linear equation set by the S-L Q iterative method is as follows:

(1) according toCalculating an initial residual r⁽⁰⁾；

Setting rho | | | r⁽⁰⁾||₂、v⁽⁰⁾＝r⁽⁰⁾/ρ、w⁽⁰⁾＝v⁽⁰⁾Setting up an initial vector ofInitial parameter settings are β⁽⁰⁾＝0、κ⁽⁰⁾＝ρ、c＝-1、ζ⁽⁰⁾＝0、g⁽⁰⁾0 andwherein | · | purple₂Represents a 2-norm, 0_K×1Is a K × 1-dimensional zero vector, superscript (·)⁽⁰⁾An initial value representing a setup iteration;

(2) the intermediate quantity is updated and the intermediate quantity is updated,

(3) updating a system of linear equationsThe vector of the solution of (a) is,

(4) update the intermediate vector, w^(t+1)＝ζ^(t+1)w^(t)+c^(t+1)v^(t+1)、

(5) Judging whether T is true or not, if so, finishing iteration and outputtingOtherwise, jumping to the step (2); after completing T iterations, the solution vector of the linear equation systemAs an estimate of the transmitted signal vector.

Technical Field

The invention belongs to the technical field of wireless communication, and relates to a large-scale MIMO system iterative signal detection method based on symmetry L Q.

Background

With the increasing number of users using the mobile internet and the internet of things, in order to meet future requirements, a fifth Generation mobile communication system (5st Generation, 5G) becomes a research hotspot, and a Multiple-Input Multiple-Output (MIMO) technology is one of the main 5G technologies.

Although the large-scale MIMO system has excellent performance, due to the large increase of the number of antennas, the application to the business still faces a great problem, such AS high computation complexity of signal detection, the maximum likelihood (M L) detection algorithm is the optimal detection algorithm, but the computation complexity of the algorithm shows an exponential growth rule with the number of users increasing, in order to reduce the computation complexity of the M L detection algorithm, the K-best detection algorithm and the SD decoding detection algorithm are proposed one after another, the K-best detection algorithm is implemented by using a breadth-first strategy, the SD decoding detection algorithm is based on a depth-first strategy, the computation complexity of the two algorithms is still very high, therefore, the objective of reducing the computation complexity can be achieved by using a local Search optimization algorithm, such AS likelihood-raising Search algorithm (L elihood-using Search, L AS) and active Tabu Search (Tabu Search, TS), the basic idea of the two algorithms is to determine the optimal vector according to the initial vector, determine the optimal AS-adjacent-receiving algorithm, the N-N.

To date, many papers have shown that when the number of base station antennas and the number of transmitting antennas satisfy the condition of N > K, N represents the number of receiving antennas at the base station, and K represents the number of users using a single antenna, the channels between users gradually tend to be orthogonal, so if the influence of additive white gaussian noise is not considered, the linear detection algorithm can also achieve the performance of M L detection algorithms, such as Zero-Forcing (ZF) detection algorithm and Minimum Mean-Square Error (MMSE) detection algorithmThe inversion operation results in a too high computational complexity of the linear detection algorithm. In order to reduce the computational complexity of the MMSE detection algorithm, Cholesky decomposition is adopted to avoid the operation of matrix inversion, but the computational complexity of the method is O (K)³) And the method is difficult to be applied to practice, so that the problem of high computational complexity of the traditional linear detection algorithm must be solved.

Disclosure of Invention

In view of this, the present invention provides a symmetric L Q-based iterative signal detection method for a large-scale MIMO system, which solves the problem of excessive computational complexity of the conventional linear detection algorithm due to the inversion operation of a high-dimensional matrix.

In order to achieve the purpose, the invention provides the following technical scheme:

a large-scale MIMO system iterative signal detection method based on symmetry L Q comprises the following steps:

s1: converting a multi-user signal recovery problem in a large-scale MIMO system into a linear equation solving system;

and S2, iteratively solving the linear equation set by adopting a Symmetric L Q (symmetry L Q, S-L Q) method, and taking a solution vector of the linear equation set as an estimated value of a transmission signal vector after the maximum iteration times are finished.

Further, the step S1 specifically includes: in an uplink multi-user large-scale MIMO system, a base station is provided with N receiving antennas to serve K single-antenna users; the base station receives the signal vector y as Hx + n by minimum mean-Square Error (MMSE), and the user transmits the estimated value of the vector xWherein F ═ H^HH+σ²I_K)^-1H^HAn MMSE equalization matrix; g is H^HH，W＝G+σ²I_K，Then satisfyThen, the problem of signal detection is converted into the solution of a linear equation set; wherein y is_N×1Is a received signal vector, x_K×1Is a transmitted signal vector, H_N×KIs a channel matrix; n is a noise vector, assumed to obey a mean of 0 and a covariance matrix of σ²I_NComplex Gaussian random variable of (I)_NIs an N × N-dimensional unit matrix;is an estimate of the transmitted signal vector, W is the MMSE filtering matrix,is a matched filtered signal; (.)^HIndicating the operation of conjugate transpose of the matrix, superscript (. cndot.)^-1Representing the inversion of the matrix.

Further, the step S2 specifically includes iteratively solving the linear equation system by using a symmetric L Q methodIn the t +1 th iteration, the solution vector is updated:

wherein, g^(t+1)、c^(t+1)And ζ^(t+1)As an iteration coefficient, w^(t)And v^(t+1)For iterative vector, superscript (. cndot.)^(t)Representing the T iteration, T ∈ 1,2, …, T, T representing the maximum iteration time, after finishing the T iterations, the linear equation set isSolution vector ofAs an estimate of the transmitted signal vector.

Further, a symmetric L Q method is used to iteratively solve the system of linear equationsThe specific process comprises the following steps:

let w and v be two sets of linearly independent vectors, let K_m＝span w，K_mIs the right subspace, L_m＝span v，L_mIs the left subspace; for linear system of equationsIterative solution is carried out to seek a value belonging to K_mApproximate solution ofThe condition of Petorv-Galerkin is satisfied:

by using the dominant diagonal dominance characteristic of the matrix W, the initial solution of the S-L Q iterative method is set asThe process of solving the linear equation set by the S-L Q iterative method is as follows:

(1) according toCalculating an initial residual r⁽⁰⁾；

Setting rho | | | r⁽⁰⁾||₂、v⁽⁰⁾＝r⁽⁰⁾/ρ、w⁽⁰⁾＝v⁽⁰⁾Setting up an initial vector ofInitial parameter settings are β⁽⁰⁾＝0、κ⁽⁰⁾＝、c＝-1、ζ⁽⁰⁾＝0、g⁽⁰⁾0 andwherein | · | purple₂Represents a 2-norm, 0_K×1Is a K × 1-dimensional zero vector, superscript (·)⁽⁰⁾An initial value representing a setup iteration;

(2) the intermediate quantity is updated and the intermediate quantity is updated,

(3) updating a system of linear equationsThe vector of the solution of (a) is,

(4) update the intermediate vector, w^(t+1)＝ζ^(t+1)w^(t)-c^(t+1)v^(t+1)、

(5) Judging whether T is true or not, if so, finishing iteration and outputtingOtherwise, jumping to the step (2); after completing T iterations, the solution vector of the linear equation systemAs an estimate of the transmitted signal vector.

The invention has the beneficial effects that: compared with the traditional linear detection algorithm, the iterative signal detection algorithm of the large-scale MIMO system avoids the operation of matrix inversion, greatly reduces the computational complexity, and can obtain the performance close to the MMSE detection algorithm after a plurality of iterative operations.

Additional advantages, objects, and features of the invention will be set forth in part in the description which follows and in part will become apparent to those having ordinary skill in the art upon examination of the following or may be learned from practice of the invention. The objectives and other advantages of the invention may be realized and attained by the means of the instrumentalities and combinations particularly pointed out hereinafter.

Drawings

For the purposes of promoting a better understanding of the objects, aspects and advantages of the invention, reference will now be made to the following detailed description taken in conjunction with the accompanying drawings in which:

FIG. 1 is a diagram of a model of a MIMO communication system;

FIG. 2 is a general flowchart of a signal detection method in a low-complexity large-scale MIMO system according to the present invention;

fig. 3 is a flowchart of a specific implementation of the iterative signal detection algorithm of the massive MIMO system based on S-L Q provided in the present invention.

Detailed Description

The embodiments of the present invention are described below with reference to specific embodiments, and other advantages and effects of the present invention will be easily understood by those skilled in the art from the disclosure of the present specification. The invention is capable of other and different embodiments and of being practiced or of being carried out in various ways, and its several details are capable of modification in various respects, all without departing from the spirit and scope of the present invention. It should be noted that the drawings provided in the following embodiments are only for illustrating the basic idea of the present invention in a schematic way, and the features in the following embodiments and examples may be combined with each other without conflict.

Referring to fig. 1 to fig. 3, a system environment according to an embodiment of the present invention is a multi-user massive MIMO system, as shown in fig. 1. Suppose a massive MIMO system is configured with N antennas and K (K < N) single-antenna user equipments at the base station end. At the transmitting end, K users transmit respective signals to be transmitted simultaneously through respective transmitting antennas. At the receiving end, the base station performs signal detection according to the received combined signal, thereby recovering the transmitted signal. Therefore, the base station received signal in the massive MIMO system can be expressed as:

y＝Hx+n

wherein y is_N×1Is a received signal vector, x_K×1Is a vector of transmitted signals, H_N×KIs a Channel matrix, and each element is independent from each other, obeys complex gaussian random variable distribution with mean value of 0 and variance of 1, and assumes the known Channel State Information (CSI) of the base station; n is a noise vector, assumed to obey a mean of 0 and a covariance matrix of σ²I_NComplex Gaussian random variable of (I)_NIs an N × N dimensional identity matrix.

Based on the system and with reference to fig. 2 and fig. 3, each step of the iterative multi-user signal detection algorithm of the large-scale MIMO system based on the symmetric L Q method specifically includes:

(1) converting the multi-user signal recovery problem into solving a system of linear equations

At the receiving end, the base station end knows the received signal vector y and the channel matrix H, and can recover the signals of multiple users by using the equalizing filter matrix of MMSE. G is H^HH，W＝G+σ²I_K，The MMSE equalization filter matrix can therefore be equivalent to: f ═ H^HH+σ²I_K)^-1H^HWhereby the user transmit vector has an estimated value ofBased on this, the target problem is converted into a solution of a system of linear equations. y is_N×1Is a received signal vector, H_N×KIs a channel matrix, n is a noise vector;is an estimate of the transmitted signal vector, W is the MMSE filtering matrix,is a matched filtered signal; (.)^HIndicating the operation of conjugate transpose of the matrix, superscript (. cndot.)^-1Representing the inversion of the matrix.

(2) Algorithm for iterative solution by adopting S-L Q method

Let w and v be two sets of linearly independent vectors, let K_m＝span w，K_mIs the right subspace, L_m＝span v，L_mIs the left subspace. For linear system of equationsIterative solution is carried out to seek a value belonging to K_mApproximate solution ofThe condition of Petorv-Galerkin is satisfied:

firstly, in order to accelerate the convergence rate of the S-L Q iterative method, the initial solution of the S-L Q iterative method can be set asThe S-L Q iterative method solves the linear equation set as follows:

a. according toCalculating an initial residual r⁽⁰⁾. Setting rho | | | r⁽⁰⁾||₂、v⁽⁰⁾＝r⁽⁰⁾/ρ、w⁽⁰⁾＝v⁽⁰⁾Setting up an initial vector ofInitial parameter settings are β⁽⁰⁾＝0、κ⁽⁰⁾＝、c＝-1、ζ⁽⁰⁾＝0、g⁽⁰⁾0 andwherein | · | purple₂Represents a 2-norm, 0_K×1Is a K × 1-dimensional zero vector, superscript (·)⁽⁰⁾An initial value representing a setup iteration;

b. the intermediate quantity is updated and the intermediate quantity is updated,

c. the solution vector of the system of linear equations is updated,the upper label (·)^(t)Represents the T-th iteration, T ∈ 1,2, …, T;

d. update the intermediate vector, w^(t+1)＝ζ^(t+1)w^(t)+c^(t+1)v^(t+1)、

e. Judging whether T is true or not, if so, finishing iteration and outputtingOtherwise, jumping to the step (2). After completing T iterations, the solution vector of the linear equation systemAs an estimate of the transmitted signal vector.

Finally, the above embodiments are only intended to illustrate the technical solutions of the present invention and not to limit the present invention, and although the present invention has been described in detail with reference to the preferred embodiments, it will be understood by those skilled in the art that modifications or equivalent substitutions may be made on the technical solutions of the present invention without departing from the spirit and scope of the technical solutions, and all of them should be covered by the claims of the present invention.