阵列信号处理基础（一、协方差矩阵）

背景

        在处理阵列信号的时候，为了获得空间信号维度的相关性，以估计目标的信息。故使用协方差矩阵能够获得这些，因为协方差矩阵是每一维度下（也就是阵元）信号的相关性。当两个维度相关时，信号的协方差也是最大的。

源代码

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
%            阵列信号基础（协方差矩阵）
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
clc;
close all;
clear all;


%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
%                  信源
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
% 随机产生10*3维的整数矩阵作为样本集，8为样本的个数，3为样本的维数
s=fix(rand(8,3)*10);  


%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
%                 法一
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
cov(s)    % 计算协方差矩阵


%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
%                 法二
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
dim1=s(:,1); %得到第1维的数据
dim2=s(:,2); %得到第2维的数据
dim3=s(:,3); %得到第3维的数据
% size(mysample,1)  %第一维的大小
cov12=sum((dim1-mean(dim1)).*(dim2-mean(dim2)))/(size(s,1)-1);  %计算dim1和dim2 的协方差
cov21=cov12;
cov13=sum((dim1-mean(dim1)).*(dim3-mean(dim3)))/(size(s,1)-1);  %计算dim1和dim3 的协方差
cov31=cov13;
cov23=sum((dim2-mean(dim2)).*(dim3-mean(dim3)))/(size(s,1)-1);  %计算dim2和dim3 的协方差
cov32=cov23;
% 协方差矩阵对角线上的元素就是各个维度的方差，下面计算
var1=std(dim1)^2;
var2=std(dim2)^2;
var3=std(dim3)^2;
answer=[var1,cov12,cov13;cov21,var2,cov23;cov31,cov32,var3]


%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
%                结论
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
% 法一直接利用函数得到，法二通过推导得到，有利于理解协方差矩阵本质
%           三维协方差矩阵
%      cov(x,x) cov(x,y) cov(x,z)
%      cov(y,x) cov(y,y) cov(y,z)
%      cov(z,x) cov(z,y) cov(z,z)

仿真结果