WO2013029226A1

WO2013029226A1 - Method and system for channel parameter estimation

Info

Publication number: WO2013029226A1
Application number: PCT/CN2011/079063
Authority: WO
Inventors: 张建华; 张平; 高新颖; 熊艳伟
Original assignee: 北京邮电大学
Priority date: 2011-08-29
Filing date: 2011-08-29
Publication date: 2013-03-07

Abstract

Disclosed are a method and system for channel parameter estimation. The method comprises the following steps: step A: initializing a channel by using a beamforming method, acquiring an initial value of a channel parameter set, and determining an initialization path count; step B: dividing a multipath package on a latency domain, performing parameter estimation on paths of the initialization path count by using a multipath package scheme on the basis of the initial value of the channel parameter set, and acquiring a parameter set of each path selected; and step C: judging, on the basis of the parameter set of the each path estimated, if the path is discarded. The method and system of the present invention allow for improved consistence with the propagation mechanism that paths exist in clusters in a broadband system, are applicable in handling mass volume of data, and improve the speed of computation.

Description

一种信道参数估计方法和*** Channel parameter estimation method and system

技术领域Technical field

本发明涉及无线通信领域，尤其涉及对无线通信信道的参数估计。 The present invention relates to the field of wireless communications, and more particularly to parameter estimation of wireless communication channels.

背景技术Background technique

移动通信的发展日新月异，从第一代蜂窝移动通信***到目前广泛研究的***高级国际移动通信（ IMT-Advanced ）***，***的传输速率不断提高，带宽不断增加，信道在时延、路径损耗、空时相关性、频率选择性等诸多方面都会表现出新的特性，这对于***设计将提出新的挑战。因此，研究无线信道传播特性是研究未来移动通信技术的前提与关键。 The development of mobile communications is changing with each passing day, from the first generation of cellular mobile communication systems to the fourth generation of advanced international mobile communications ( IMT-Advanced The system, the transmission rate of the system continues to increase, the bandwidth continues to increase, and the channel will exhibit new characteristics in terms of delay, path loss, space-time correlation, frequency selectivity, etc., which will pose new challenges for system design. Therefore, studying the wireless channel propagation characteristics is the premise and key to study the future mobile communication technology.

在实际的各种地理环境中进行无线信道的测量工作是了解真实信道特征的最佳途径，被世界各国广泛采用。从信道测量结果中进行参数估计是分析信道特性的关键步骤。准确的信道测量与参数估计能够准确反映真实传播环境中的信道特性，从而可以建立合适的信道模型，为无线通信***中的传输技术、资源管理和网络规划等研究提供参考和依据。 The measurement of wireless channels in actual geographical environments is the best way to understand the characteristics of real channels, and is widely used in countries around the world. Parameter estimation from channel measurements is a critical step in analyzing channel characteristics. Accurate channel measurement and parameter estimation can accurately reflect the channel characteristics in the real propagation environment, so that a suitable channel model can be established, which provides reference and basis for the research of transmission technology, resource management and network planning in wireless communication systems.

技术问题technical problem

目前，针对宽带信道参数估计，常用的方法有：波束成形方法、子空间法及由最大似然估计准则衍生出来的算法：期望最大化算法（ EM ）和空间交替广义期望最大化算法（ SAGE ）。以上各种方法都有各自的优缺点，例如子空间法的运算复杂度要比波束成形方法大， SAGE 算法对弱径的估计能力比较弱等。 At present, the commonly used methods for wideband channel parameter estimation are: beamforming method, subspace method and algorithms derived from maximum likelihood estimation criteria: expectation maximization algorithm (EM) and spatial alternating generalized expectation maximization algorithm ( SAGE). Each of the above methods has its own advantages and disadvantages. For example, the computational complexity of the subspace method is larger than that of the beamforming method, and the SAGE algorithm has a weaker estimation ability for the weak path.

技术解决方案Technical solution

为了解决上述参数估计中存在的问题，更加准确、有效地进行参数估计，本发明提供了一种信道参数估计方法，利用波束成形方法进行初始化，径参数估计主循环采用期望最大化（EM）算法。 In order to solve the problems in the above parameter estimation and perform parameter estimation more accurately and effectively, the present invention provides a The channel parameter estimation method is initialized by a beamforming method, and the main parameter of the path parameter estimation adopts an expectation maximization (EM) algorithm.

本发明提供了一种信道参数估计方法，其特征在于，包括以下步骤：步骤 A ：利用波束成形法对信道进行初始化，获得信道参数集的初始值，确定初始化的径数量；步骤 B ：在时延域上划分多径包，基于所述信道参数集的初始值，以多径包的方式对所述初始化径数量的径进行参数估计，得到所选取的每条径的参数集；步骤 C ：基于所估计的所述每条径的参数集，判断该径是否被舍弃。 The invention provides a channel parameter estimation method, which comprises the following steps: Step A : Initializing the channel by beamforming, obtaining the initial value of the channel parameter set, and determining the number of initialized paths; Step B Dividing a multipath packet on a delay domain, and estimating a path of the number of initialization paths by using a multipath packet based on an initial value of the channel parameter set, to obtain a parameter set of each path selected; Step C : determining whether the path is discarded based on the estimated parameter set of each path.

进一步，该方法还包括，步骤 A 中的初始化具体包括：根据测量得到的信道冲击响应矩阵计算空间相关矩阵，根据完全空间相关矩阵得到双角度功率谱，然后依次得到离开角的水平角、离开角的垂直角、到达角的水平角和到达角的垂直角的估计值、多普勒初始值、极化复幅度矩阵。进一步，该方法还包括，在步骤 A 中，采用 Bartlett 波束成形法根据空间相关矩阵确定角度功率谱。进一步，该方法还包括，所述步骤 A 中，信道参数集的初始值包括：径时延值、离开角的水平角、离开角的垂直角、到达角的水平角、到达角的垂直角、多普勒值和 / 或极化复幅度矩阵。进一步，该方法还包括，所述步骤 A 中，所述确定初始化的径数量具体包括：确定信道的径总数，将所有径按功率从大到小进行排序，选取预定数量的径作为初始化的径。进一步，该方法还包括，所述步骤 B 中，对径参数进行估计具体包括：基于所述信道参数集的初始值和完备信号的条件期望值，以划分多径包的方式通过迭代方式估计每条径的参数。进一步，该方法还包括，通过对完备信号的似然函数求最大值对每条径的参数进行估计，将似然函数为最大值时的径参数作为每条径的参数估计值。进一步，该方法还包括，所述步骤 C 中，所述预设条件为：计算每条径功率，将径功率低于功率延时阈值的径舍弃；和 / 或，为计算天线阵元对每条径产生的增益，将增益小于径平均增益预定比例的径舍弃。进一步，该方法还包括，采用期望最大化方法进行径的参数估计。 Further, the method further includes step A The initialization includes: calculating a spatial correlation matrix according to the measured channel impulse response matrix, obtaining a double angle power spectrum according to the complete spatial correlation matrix, and then sequentially obtaining a horizontal angle of the exit angle, a vertical angle of the exit angle, and a horizontal angle of the arrival angle. And the estimated value of the vertical angle of the angle of arrival, the Doppler initial value, and the polarization complex amplitude matrix. Further, the method further comprises, in step A, determining an angular power spectrum from the spatial correlation matrix using Bartlett beamforming. Further, the method further includes the step A The initial values of the channel parameter set include: a path delay value, a horizontal angle of the exit angle, a vertical angle of the exit angle, a horizontal angle of the arrival angle, a vertical angle of the arrival angle, a Doppler value, and/or a polarization complex amplitude. matrix. Further, the method further includes the step A The determining the number of initialized paths specifically includes: determining a total number of paths of the channels, sorting all the paths by power from large to small, and selecting a predetermined number of paths as the initialized path. Further, the method further includes the step B The estimating the path parameter specifically includes: estimating the parameter of each path in an iterative manner by dividing the multipath packet based on the initial value of the channel parameter set and the conditional expectation value of the complete signal. Further, the method further comprises: estimating the parameter of each path by obtaining a maximum value of the likelihood function of the complete signal, and using the path parameter when the likelihood function is the maximum value as the parameter estimation value of each path. Further, the method further includes the step In C, the preset condition is: calculating each path power, discarding the path whose path power is lower than the power delay threshold; and/or, in order to calculate the gain generated by the antenna element for each path, the gain is smaller than the diameter The average gain is a predetermined percentage of the diameter discarded. Further, the method further comprises performing parameter estimation of the path using the expectation maximization method.

本发明还提供了一种信道参数估计***，其特征在于，包括以下单元：初始化单元：其利用波束成形法对信道进行初始化，获得信道参数集的初始值，确定初始化的径数量；参数估计单元：其在时延域上划分多径包，基于所述信道参数集的初始值，以多径包的方式对所述初始化径数量的径进行参数估计，得到所选取的每条径的参数集；有效性验证单元：其基于所估计的所述每条径的参数集，判断该径是否被舍弃。The present invention also provides a channel parameter estimation system, comprising: an initialization unit: initializing a channel by using a beamforming method, obtaining an initial value of a channel parameter set, determining an initial number of paths; and a parameter estimation unit It divides the multipath packet on the delay domain, and based on the initial value of the channel parameter set, performs parameter estimation on the path of the number of initialization paths in a multipath packet manner, and obtains a parameter set of each path selected. Validity verification unit: based on the estimated parameter set of each path, it is determined whether the path is discarded.

有益效果Beneficial effect

与现有技术相比，本发明具有以下优点：本发明公开的信道参数估计方法和***在初始化时采用了波束成形的方法，提高了对弱径的估计能力，更符合宽带***中径以簇存在的传播机理。主循环时采用期望最大化方法，适用于处理大量的数据。通过划分多径包的方式对经参数并行估计，提高了运算速率。本发明还进一步对搜索到的径的进行了有效性验证。Compared with the prior art, the present invention has the following advantages: The channel parameter estimation method and system disclosed by the invention adopts a beamforming method in initialization, which improves the estimation ability of the weak path, and is more in line with the propagation mechanism of the cluster in the broadband system. The main loop uses the expectation maximization method, which is suitable for processing large amounts of data. Parallel estimation of the parameters by dividing the multipath packet improves the operation rate. The present invention further validates the searched path.

本发明的其它特征和优点将在随后的说明书中阐述，并且，部分地从说明书中变得显而易见，或者通过实施本发明而了解。本发明的目的和其他优点可通过在说明书、权利要求书以及附图中所特别指出的结构来实现和获得。Other features and advantages of the invention will be set forth in the description which follows, The objectives and other advantages of the invention may be realized and obtained by means of the structure particularly pointed in the appended claims.

附图说明DRAWINGS

附图用来提供对本发明的进一步理解，并且构成说明书的一部分，与本发明的实施例一起用于解释本发明，并不构成对本发明的限制。在附图中：The drawings are intended to provide a further understanding of the invention, and are intended to be a In the drawing:

图 1 是根据本发明实施例一的信道参数估计方法的流程图；1 is a flowchart of a channel parameter estimation method according to Embodiment 1 of the present invention;

图 2 是根据本发明实施例一的对信道参数进行初始化的流程图；2 is a flowchart of initializing channel parameters according to Embodiment 1 of the present invention;

图 3 是根据本发明实施例一的信道参数估计的主循环流程示意图；3 is a schematic diagram of a main loop process of channel parameter estimation according to Embodiment 1 of the present invention;

图 4 是根据实施例一的信道参数估计方法的径有效性验证流程示意图；4 is a schematic flowchart of a path validity verification process of a channel parameter estimation method according to Embodiment 1;

图 5 是根据本发明实施例一的在功率延时谱上划分多径包的示意图；FIG. 5 is a schematic diagram of dividing a multipath packet on a power delay spectrum according to Embodiment 1 of the present invention; FIG.

图 6 是根据本发明实施例二的信道参数估计***的结构图。6 is a structural diagram of a channel parameter estimation system according to a second embodiment of the present invention.

本发明的实施方式Embodiments of the invention

以下将结合附图及实施例来详细说明本发明的实施方式，借此对本发明如何应用技术手段来解决技术问题，并达成技术效果的实现过程能充分理解并据以实施。需要说明的是，只要不构成冲突，本发明中的各个实施例以及各实施例中的各个特征可以相互结合，所形成的技术方案均在本发明的保护范围之内。 The embodiments of the present invention will be described in detail below with reference to the accompanying drawings and embodiments, in which the present invention can be applied to the technical problems, and the implementation of the technical effects can be fully understood and implemented. It should be noted that the various embodiments of the present invention and the various features of the various embodiments may be combined with each other, and the technical solutions formed are all within the scope of the present invention.

另外，在附图的流程图示出的步骤可以在诸如一组计算机可执行指令的计算机***中执行，并且，虽然在流程图中示出了逻辑顺序，但是在某些情况下，可以以不同于此处的顺序执行所示出或描述的步骤。 Additionally, the steps illustrated in the flowchart of the figures may be executed in a computer system such as a set of computer executable instructions, and although the logical order is illustrated in the flowchart, in some cases may be different The steps shown or described are performed in the order herein.

实施例一 Embodiment 1

图 1 是根据本发明实施例一的信道参数估计方法的流程图，下面参照图 1 详细说明该方法的各步骤。步骤 S110 ，利用波束成形法对信道参数进行初始化，确定初始化的径数量。本步骤对信道的径进行初始化，得到径的参数的初始值。根据测量得到的信道冲击响应矩阵计算（完全）空间相关矩阵，得到双角度功率谱，然后依次得到四个角度（离开角的水平角、离开角的垂直角、到达角的水平角和到达角的垂直角）的估计值、多普勒初始值、极化复幅度矩阵，最终得到所有参数的初始值，形成初始参数集。同时，也对径的总数进行初始化。 1 is a flowchart of a channel parameter estimation method according to Embodiment 1 of the present invention, and the steps of the method will be described in detail below with reference to FIG. step S110: Initialize channel parameters by using a beamforming method to determine the number of initialized paths. This step is for the channel The path is initialized to obtain the initial value of the path parameter. Calculating the (complete) spatial correlation matrix from the measured channel impulse response matrix, obtaining a double-angle power spectrum, and then sequentially obtaining four angles (the horizontal angle of the exit angle, the vertical angle of the exit angle, the horizontal angle of the arrival angle, and the angle of arrival) The estimated value of the vertical angle, the Doppler initial value, and the polarization complex amplitude matrix, and finally the initial values of all the parameters are obtained to form an initial parameter set. At the same time, the total number of paths is also initialized.

图2根据本发明实施例一的对信道参数进行初始化的流程图。下面参照图2具体说明初始化的各步骤：步骤1110，根据信道矩阵定义空间相关矩阵。在不同的输入输出***中，相关矩阵的表达方式也不同，具体公式如下：在多入多出（ MIMO ）***中，空间相关矩阵 R_H 为：

其中，I表示在第k个时延 τ _k 上信道矩阵的采样点数， H_i ( τ _k ) 表示第 τ _k 个时延上第i个采样点的信道矩阵， vec(·) 表示矩阵按列向量化， ( ·) ^H 表示共轭转置。在多入单出（ MISO ）***中，发送端空间相关矩阵 R 为：

其中 h _i 表示在时延 τ _k 上第i个采样点的信道矩阵（向量）。单入多出（SIMO）***中，接收端空间相关矩阵 R 为：

2 is a flow chart for initializing channel parameters according to a first embodiment of the present invention. The steps of initialization are specifically described below with reference to FIG. 2: Step 1110, a spatial correlation matrix is defined according to a channel matrix. In different input and output systems, the expression matrix is expressed differently. The specific formula is as follows: In a multiple-input multiple-output (MIMO) system, the spatial correlation matrix R _H is:

Wherein, I represents the number of sampling points of the channel matrix on the kth delay τ _k , H _i ( τ _k ) represents the channel matrix of the ith sampling point on the τ _kth delay, and vec(·) represents the matrix by column Vectorization, ( ·) ^H represents conjugate transpose. In a Multiple Entry and Single Output (MISO) system, the spatial correlation matrix R of the sender is:

Where h _i represents the channel matrix (vector) of the ith sample point on the delay τ _k . In the single-input multiple-output (SIMO) system, the spatial correlation matrix R at the receiving end is:

步骤 1120 ，根据得到的空间相关矩阵，利用 Bartlett （巴特利特）波束成形法得到双角度功率谱，其中的双角度指离开角和到达角。在 MIMO ***中，公式如下：

其中 Ω₁，Ω₂为归一化向量，均可用下式表示： Ω_k=[cos( φ _k )sin( θ _k ),sin( φ _k )sin( θ _k ),cos( θ _k )] (k=1,2), 其中，1和2表示发送和接收天线， φ _k , θ _k 分别为水平角和垂直角；

表示克罗内克积；

其中 c_1,1(Ω₁),c_1,2(Ω₁), c_2,1(Ω₂),c_2,2(Ω₂) 为四个不同极化天线响应矢量，其中第一个下标1和2表示发送或接收天线，第二个下标1和2表示水平或垂直。MISO ***中，在发送端进行单边的波束成形：

SIMO ***中，在接收端进行单边的波束成形：

Step 1120, according to the obtained spatial correlation matrix, a double angle power spectrum is obtained by Bartlett beamforming method, wherein the double angle refers to a departure angle and an arrival angle . In a MIMO system, the formula is as follows:

Where Ω ₁ and Ω ₂ are normalized vectors, which can be expressed by the following equation: Ω _k =[cos( φ _k )sin( θ _k ), sin( φ _k )sin( θ _k ),cos( θ _k )] ( k =1, 2), where 1 and 2 represent transmitting and receiving antennas, and φ _k and θ _k are horizontal and vertical angles, respectively;

Expressing Kronecker's product;

Where c _1,1 (Ω ₁ ), c _1,2 (Ω ₁ ), c _2,1 (Ω ₂ ), c _2,2 (Ω ₂ ) are four different polarized antenna response vectors, the first of which Subscripts 1 and 2 indicate transmit or receive antennas, and second subscripts 1 and 2 indicate horizontal or vertical. In the MISO system, one-sided beamforming is performed at the transmitting end:

In the SIMO system, one-sided beamforming is performed at the receiving end:

步骤 1130 ，求取使角度功率谱达到极大值的四个角度。在 MIMO ***中：

Step 1130, obtaining four angles that cause the angular power spectrum to reach a maximum value. In a MIMO system:

在 MISO ***中：

In the MISO system:

在 SIMO ***中：

其中， φ _1,k ⁽⁰⁾ , θ _1,k ⁽⁰⁾ , φ _2,k ⁽⁰⁾ , θ _2,k ⁽⁰⁾ 分别表示离开角的水平角、离开角的垂直角、到达角的水平角、到达角的垂直角。 In the SIMO system:

Where φ _1,k ⁽⁰⁾ , θ _1,k ⁽⁰⁾ , φ _2,k ⁽⁰⁾ , θ _2,k ⁽⁰⁾ represent the horizontal angle of the exit angle, the vertical angle of the exit angle, and the angle of arrival, respectively Horizontal angle, vertical angle of arrival angle.

步骤 1140 ，计算得到多普勒初始值。多普勒初始值计算公式如下：

其中，

表示各归一化的天线响应矢量，M为发送天线数，N为接收天线数，y (t)为接收端观测到的数据，即信道冲击响应，(·)^*表示伴随矩阵，j是复数单位，t _i,n,m表示第 m 个发送天线和第 n 个接收天线之间第 i 个信道样本的起始时间，u是发送的信号，T _sc 是接收端每个码的持续时间，τ _k ⁰⁾表示延时的初始估计值。 In step 1140, the initial Doppler value is calculated. The Doppler initial value is calculated as follows:

among them,

Represents each normalized antenna response vector, M is the number of transmitting antennas, N is the number of receiving antennas, y ( t ) is the data observed by the receiving end, that is, the channel impulse response, (·) ^* indicates the adjoint matrix, and j is the complex number The unit, t _i,n,m represents the start time of the ith channel sample between the mth transmit antenna and the nth receive antenna, u is the transmitted signal, and T _sc is the duration of each code at the receiving end. τ _k ⁰⁾ represents the initial estimate of the delay.

步骤 1150 ，计算得到极化复幅度矩阵。极化复幅度矩阵的计算公式如下：

其中 P 为发送信号功率， D (Ω_2,k,Ω_1,k) 的表达式为：

其中， C_k(Ω)=[c_k _,1(Ω),c_k _,2(Ω)] (k=1,2) ，表示由 c_k _,1(Ω),c_k _,2(Ω) 所组合的向量， Ω_2,k, Ω_1,k 分别为在时延 τ_k 上的归一化向量。

其中 X_k( τ_k,f _d,k ) 是一个基于 TDM-MIMO 结构的 M×N 维矩阵，表达式为：

于是得到了所有参数的初始值，构成初始参数集 β _k ⁽⁰⁾ ，表达式为：

In step 1150, a polarization complex amplitude matrix is calculated. The calculation formula of the polarization complex amplitude matrix is as follows:

Where P is the transmitted signal power, and the expression D (Ω _{2, k} , Ω _{1, k} ) is:

Where C _k (Ω) = [c _k _{, 1} (Ω), c _k _{, 2} (Ω)] ( k =1, 2) , denoted by c _k _{, 1} (Ω), c _k _{, 2} (Ω) The combined vectors, Ω _{2, k} , Ω _{1, k} are the normalized vectors at time delay τ _k _, respectively.

Where X _k ( τ _k , f _d,k ) is an M×N dimensional matrix based on the TDM-MIMO structure, and the expression is:

Then the initial values of all the parameters are obtained, which constitutes the initial parameter set β _k ⁽⁰⁾ , and the expression is:

步骤 1160 ，初始化径数量。在参数初始化的同时对径总数也进行初始化，具体的，在每个时延 τ_k (k=1,…,K) 上取空间谱上 ε 个极大值对应的径参数，那么径总数为 εK ，对这些初始径按功率，由大到小进行排序，其中功率是按照测量数据得到的信道冲击响应的模的平方得到的，选出前 L 条径作为最终的初始化径数量的结果，表达式如下：

以上为步骤 S110 对信道参数进行初始化的各个子步骤，经过步骤S110的各个子步骤后，得到符合要求的 L 条径以及这 L 条径的初始值。In step 1160, the number of paths is initialized. The total number of paths is also initialized at the same time as the parameter initialization. Specifically, on each time delay τ _k ( k =1,..., K ), the path parameters corresponding to ε maxima in the spatial spectrum are taken, then the total number of paths is εK , the power of these initial diameters is sorted from large to small, wherein the power is obtained by the square of the mode of the channel impulse response obtained from the measured data, and the front L path is selected as the result of the final number of initialization paths, and the expression is expressed. The formula is as follows:

In the above substeps of initializing the channel parameters in step S110, after the sub-steps of step S110, the L-paths satisfying the requirements and the initial values of the L-paths are obtained.

步骤S120，以划分多径包的方式对选取的径进行参数估计，得到每条径的参数集。在本步骤中，依据先前步骤获得的每条径的初始值，通过循环运算最后得出每条径的参数估计值，作为每条径的参数集。其中循环运算的终止条件可设为：当本次得到的参数估计与上次得到的相同则循环结束。也可设定最大循环次数以防止进入无限循环。图3为根据本实施例的信道参数估计的主循环流程示意图。下面参照图 3 说明主循环流程各步骤，主循环流程具体包括以下步骤：

Step S120: Perform parameter estimation on the selected path by dividing the multipath packet to obtain a parameter set of each path. In this step, according to the initial value of each path obtained in the previous step, the parameter estimation value of each path is finally obtained by the loop operation as the parameter set of each path. The termination condition of the loop operation may be set to: when the parameter estimation obtained this time is the same as that obtained last time, the loop ends. You can also set the maximum number of cycles to prevent entry into an infinite loop. FIG. 3 is a schematic diagram of a main loop flow of channel parameter estimation according to the present embodiment. The steps of the main loop process are described below with reference to FIG. 3. The main loop process specifically includes the following steps:

步骤 1220 ，划分多径包。如图 5 所示，将测量得到的信道冲击响应作为测量数据，对同一处多点采样的测量数据取平均值，根据该测量数据的平均值得到功率延时谱 PDP ，在功率时延阈值 P th （该功率时延阈值是功率的界限值，在这个界限以下的数据可认为是无效数据）之上寻找局部极大值点作为多径包的峰值，在局部极大值的左右寻找到局部极小值，用区间 [τ _q _,1, τ _q _,2 ] 表示第 q 个多径包的持续时间，如果相邻多径包的时延界是相同的，则 τ _q _,2 =τ _q _+1,1 。图 5 所示的 PDP 中划分了 5 个多径包。 In step 1220, the multipath packet is divided. As shown in FIG. 5, the measured channel impulse response is used as measurement data, and the measurement data of the same multi-point sampling is averaged, and the power delay spectrum PDP is obtained according to the average value of the measurement data, and the power delay threshold P is obtained. Th (the power delay threshold is the limit value of the power, the data below this limit can be considered as invalid data). The local maximum point is found as the peak of the multipath packet, and the local maximum is found locally. For the minimum value, the interval [ τ _q _,1 , τ _q _,2 ] is used to represent the duration of the qth multipath packet. If the delay bounds of adjacent multipath packets are the same, then τ _q _{, 2} = τ _q _+1,1 . Five multipath packets are divided into the PDP shown in Figure 5.

步骤 1230 ，以划分多径包的方式，根据先前求得的条件期望值，并行地对完备信号的似然函数求最大值，得出多径包中各条径的参数的估计值，具体公式如下：

通过以下两个公式求得：

得到各个参数的估计值。判断是否满足运算结束条件，若不满足，则将得到的估计值作为下次迭代运算的输入，若满足，将参数的估计值作为本步骤的最终运算结果 β_l。通过多径包并行对参数进行估计，可提高计算速率。Step 1230, in the manner of dividing the multipath packet, according to the previously obtained conditional expectation value, the maximum value of the likelihood function of the complete signal is obtained in parallel, and the estimated value of each path parameter in the multipath packet is obtained. The specific formula is as follows: :

It is obtained by the following two formulas:

Get an estimate of each parameter. It is judged whether or not the calculation end condition is satisfied. If it is not satisfied, the obtained estimated value is used as an input of the next iteration operation, and if it is satisfied, the estimated value of the parameter is taken as the final calculation result β _{l of} this step. The parameters can be estimated by multipath packets in parallel to increase the calculation rate.

步骤 130 ，基于预设条件和所估计的每条径的参数集来判断该径是否被舍弃。在本步骤中，对径的有效性进行验证，具体地，根据先前步骤得到的对每条径估计的参数集，来判断每条径是否符合预先设定的条件，对不符合预设条件的径进行舍弃。图 4 为根据本实施例的信道参数估计方法的径有效性验证流程示意图，下面参照图 4 具体说明有效性验证的各步骤： Step 130 And determining whether the path is discarded based on the preset condition and the estimated parameter set of each path. In this step, the validity of the path is verified. Specifically, according to the parameter set estimated for each path obtained in the previous step, it is determined whether each path meets a preset condition, and the condition that does not meet the preset condition is met. The path is discarded. 4 is a schematic diagram of a path validity verification process of the channel parameter estimation method according to the present embodiment, and the steps of validity verification are specifically described below with reference to FIG. 4:

步骤 1310 ，将步骤 S120 中获得多径包中的每条径的参数集在时延域上进行信号重构，得到每条径对应的信号冲击响应记为

具体公式如下：

其中 q₁(t),q₂(t) 分别是发送天线和接收天线的天线 pattern 选择矩阵， A_l 是第 l 条径的复幅度矩阵。根据信号冲击响应计算径功率，具体地，径功率为每条径对应的冲击响应的模的平方，然后将功率低于功率时延阈值 P _th 的径被舍弃。 Step 1310, step S120 The parameter set of each path in the multipath packet is obtained by performing signal reconstruction on the delay domain, and the signal impulse response corresponding to each path is recorded as

The specific formula is as follows:

Where q₁(t),q₂(t) The antenna pattern selection matrix of the transmitting antenna and the receiving antenna, respectively, A_l Is the firstl The complex amplitude matrix of the path. Calculating the radial power according to the signal impulse response. Specifically, the radial power is the square of the mode of the impulse response corresponding to each path, and then the power is lower than the power delay threshold. P _Th The path was abandoned.

步骤 1320 ，在空域上，计算天线阵元对每条径产生的增益 WG _l ，公式如下： WG _l =E[C₂(Ω_{2, l} )C₁(Ω_{1, l} )^T] (23) 其中， C₂(Ω_{2, l} )C₁(Ω_{1, l} )^T 计算得到的结果为矩阵或向量，表示一个天线上每个阵元产生的增益， E 表示将其中的每个阵元产生的增益进行平均得到天线对上每条径的增益，将低于天线平均增益值预定比例的径舍弃，预定比例优选为 1/3 。其中，当天线的型号选定后天线的平均增益就可确定，天线一般配置有自身的***校准文件，因此天线平均增益与测量结果无关，属于已知的***自身参数。其中，步骤 1310 和步骤 1320 可择其一执行，也可相继执行以验证径的有效性。本实施例根据以上各步骤，对信道参数进行了估计，并对信道的径进行了有效性验证。Step 1320: Calculate the gain WG _l generated by the antenna element for each path in the airspace, and the formula is as follows: WG _l = E [C ₂ (Ω _{2, l} ) C ₁ (Ω _{1, l} ) ^T ] (23) Where C ₂ (Ω _{2, l} )C ₁ (Ω _{1, l} ) ^{T is} calculated as a matrix or a vector, representing the gain produced by each array element on an antenna, and E is the result of generating each of the array elements. The gain is averaged to obtain the gain of each path on the pair of antennas, and the diameter of the antenna is lower than a predetermined ratio of the average gain value, and the predetermined ratio is preferably 1/3. Wherein, the average gain of the antenna can be determined when the model of the antenna is selected, and the antenna is generally configured with its own system calibration file, so the average antenna gain is independent of the measurement result and belongs to the known system self-parameter. Step 1310 and step 1320 may be performed one by one, or may be performed sequentially to verify the validity of the path. In this embodiment, according to the above steps, the channel parameters are estimated, and the path diameter is validated.

实施例二 Embodiment 2

图 6 为根据本发明实施例二的信道参数估计***的结构图，下面根据图 6 详细说明该***的结构。 6 is a structural diagram of a channel parameter estimation system according to Embodiment 2 of the present invention, which is based on FIG. 6 The structure of the system will be described in detail.

初始化单元：其利用波束成形法对信道进行初始化，获得信道参数集的初始值，确定初始化的径数量；参数估计单元：其在时延域上划分多径包，基于所述信道参数集的初始值，以多径包的方式对所述初始化径数量的径进行参数估计，得到所选取的每条径的参数集； Initializing unit : It initializes the channel by beamforming, obtains an initial value of the channel parameter set, and determines the number of initialized paths; and the parameter estimating unit : divides the multipath packet in the delay domain, based on the initial of the channel parameter set a value, the parameters of the number of initialization paths are parameterized in a multipath packet manner, and a parameter set of each selected path is obtained;

有效性判断单元：其基于所估计的所述每条径的参数集，判断该径是否被舍弃。本实施例所描述的***中的各单元还可以用于实现实施例一中各步骤的子步骤或实施例一的各优选方案，也还可以包括实现实施例一各步骤的子步骤或优选方案的子单元，在此不作详述。 The validity judging unit : judges whether the path is discarded based on the estimated parameter set of each of the paths. The units in the system described in this embodiment may also be used to implement the sub-steps of the steps in the first embodiment or the preferred embodiments of the first embodiment, and may also include sub-steps or preferred solutions for implementing the steps of the first embodiment. Subunits are not described here.

本领域的技术人员应该明白，上述的本发明的各模块或各步骤可以用通用的计算装置来实现，它们可以集中在单个的计算装置上，或者分布在多个计算装置所组成的网络上，可选地，它们可以用计算装置可执行的程序代码来实现，从而，可以将它们存储在存储装置中由计算装置来执行，或者将它们分别制作成各个集成电路模块，或者将它们中的多个模块或步骤制作成单个集成电路模块来实现。这样，本发明不限制于任何特定的硬件和软件结合。虽然本发明所揭露的实施方式如上，但所述的内容只是为了便于理解本发明而采用的实施方式，并非用以限定本发明。任何本发明所属技术领域内的技术人员，在不脱离本发明所揭露的精神和范围的前提下，可以在实施的形式上及细节上作任何的修改与变化，但本发明的专利保护范围，仍须以所附的权利要求书所界定的范围为准。 It will be apparent to those skilled in the art that the various modules or steps of the present invention described above can be implemented by a general-purpose computing device, which can be centralized on a single computing device or distributed over a network of multiple computing devices. Alternatively, they may be implemented by program code executable by the computing device, such that they may be stored in a storage device by a computing device, or they may be fabricated into individual integrated circuit modules, or many of them Modules or steps are made into a single integrated circuit module. Thus, the invention is not limited to any specific combination of hardware and software. While the embodiments of the present invention have been described above, the described embodiments are merely illustrative of the embodiments of the invention, and are not intended to limit the invention. Any modification and variation of the form and details of the invention may be made by those skilled in the art without departing from the spirit and scope of the invention. It is still subject to the scope defined by the appended claims.

Claims

一种信道参数估计方法，其特征在于，包括以下步骤： A channel parameter estimation method, comprising the steps of:

步骤 A ：利用波束成形法对信道进行初始化，获得信道参数集的初始值，确定初始化的径数量；Step A: Initializing the channel by using a beamforming method, obtaining an initial value of the channel parameter set, and determining the number of initialized paths;

步骤 B ：在时延域上划分多径包，基于所述信道参数集的初始值，以多径包的方式对所述初始化径数量的径进行参数估计，得到所选取的每条径的参数集；Step B Dividing a multipath packet on a delay domain, and estimating a path of the number of initialization paths by using a multipath packet based on an initial value of the channel parameter set, to obtain a parameter set of each path selected;

步骤 C ：基于所估计的所述每条径的参数集，判断该径是否被舍弃。 Step C: determining whether the path is discarded based on the estimated parameter set of each path.
根据权利要求 1 所述的方法，其特征在于，所述步骤 A 中的初始化具体包括：The method according to claim 1, wherein the initializing in the step A comprises:

根据测量得到的信道冲击响应矩阵计算空间相关矩阵，根据完全空间相关矩阵得到双角度功率谱，然后依次得到离开角的水平角、离开角的垂直角、到达角的水平角和到达角的垂直角的估计值、多普勒初始值、极化复幅度矩阵。 Calculate the spatial correlation matrix according to the measured channel impulse response matrix, obtain the double angle power spectrum according to the complete spatial correlation matrix, and then obtain the horizontal angle of the exit angle, the vertical angle of the exit angle, the horizontal angle of the arrival angle and the vertical angle of the arrival angle. Estimate, Doppler initial value, polarization complex amplitude matrix.
根据权利要求2所述的方法，其特征在于，在所述步骤 A 中，采用 Bartlett 波束成形法根据空间相关矩阵得到双角度功率谱。 The method according to claim 2, characterized in that in said step A, Bartlett is employed The beamforming method obtains a dual angle power spectrum based on a spatial correlation matrix.
根据权利要求2所述的方法，其特征在于，所述步骤A中，信道参数集的初始值包括：径时延值、离开角的水平角、离开角的垂直角、到达角的水平角、到达角的垂直角、多普勒值和/或极化复幅度矩阵。 The method according to claim 2, wherein in the step A, the initial values of the channel parameter set comprise: The path delay value, the horizontal angle of the exit angle, the vertical angle of the exit angle, the horizontal angle of the angle of arrival, the vertical angle of the angle of arrival, the Doppler value, and/or the polarization complex amplitude matrix.
根据权利要求1所述的方法，其特征在于，所述步骤A中，所述确定初始化的径数量具体包括：确定信道的径总数，将所有径按功率从大到小进行排序，选取预定数量的径作为初始化的径。 The method of claim 1 wherein said determining in said step A The number of initialized paths specifically includes: determining the total number of paths of the channel, sorting all the paths by power from large to small, and selecting a predetermined number of paths as the initialized path.
根据权利要求 1 所述的方法，其特征在于，所述步骤 B 中，对径参数进行估计具体包括：基于所述信道参数集的初始值和完备信号的条件期望值，以划分多径包的方式通过迭代方式估计每条径的参数。 The method of claim 1 wherein said step B The estimating the path parameter specifically includes: estimating the parameter of each path in an iterative manner by dividing the multipath packet based on the initial value of the channel parameter set and the conditional expectation value of the complete signal.
根据权利要求 6 所述的方法，其特征在于，通过对完备信号的似然函数求最大值对每条径的参数进行估计，将似然函数为最大值时的径参数作为每条径的参数估计值。According to claim 6 The method is characterized in that the parameter of each path is estimated by finding a maximum value of the likelihood function of the complete signal, and the path parameter when the likelihood function is the maximum value is used as a parameter estimation value for each path.
根据权利要求 1 所述的方法，其特征在于，所述步骤 C 中，所述预设条件为：计算每条径功率，将径功率低于功率延时阈值的径舍弃； The method of claim 1 wherein said step C The preset condition is: calculating the power of each path, and discarding the path whose path power is lower than the power delay threshold;

和 / 或，为计算天线阵元对每条径产生的增益，将增益小于径平均增益预定比例的径舍弃。 And / or, in order to calculate the gain generated by the antenna element for each path, discard the path with a gain less than a predetermined ratio of the average diameter gain.
根据权利要求 1 所述的方法，在所述步骤 B 中，采用期望最大化方法进行径的参数估计。 The method of claim 1 wherein in step B, parameter estimation of the diameter is performed using a desired maximization method.
一种信道参数估计***，其特征在于，包括以下单元：A channel parameter estimation system, comprising the following units:

初始化单元：其利用波束成形法对信道进行初始化，获得信道参数集的初始值，确定初始化的径数量；Initializing unit: it initializes the channel by using beamforming method, obtains an initial value of the channel parameter set, and determines the number of initialized paths;

参数估计单元：其在时延域上划分多径包，基于所述信道参数集的初始值，以多径包的方式对所述初始化径数量的径进行参数估计，得到所选取的每条径的参数集；a parameter estimation unit: dividing a multipath packet on a delay domain, and estimating a path of the number of initialization paths in a multipath packet manner based on an initial value of the channel parameter set, to obtain each path selected Parameter set

有效性验证单元：其基于所估计的所述每条径的参数集，判断该径是否被舍弃。Validity verification unit: based on the estimated parameter set of each path, it is determined whether the path is discarded.