GSC.rar_gsc_gsc阻塞矩阵_广义旁瓣对消_旁瓣对消

%Course_Project clc; clear all; %% 定义信号、干扰、噪声 j = sqrt(-1); M = 8; %接收天线个数 N = 10000; %信号与干扰长度 SNR = 2; %信号和干扰的信噪比 pow_n = 1; %噪声功率 pow_s = pow_n*10.^(0.1*SNR); %信号功率 S = 2*randi([0,1],4,N)-1; %信号与干扰 正负1的随机序列 S = sqrt(pow_s) * S; %入射信号 noise = sqrt(pow_n*0.5)*(randn(M,N)+j*randn(M,N)); %复加性噪声 %% 定义信号、干扰的入射角度，写出导向矢量 theta = [10 -10 40 60]*pi/180; %入射信号的角度 第一路为感兴趣信号 其他为干扰 d = [0:M-1]; %以第一根天线为基准，定义方向矢量 A = exp(-j*pi*d'*sin(theta)); %定义阵列响应向量矩阵 y = A*S+noise; %阵列天线的输入信号 C = exp(-j*pi*d*sin(theta(1))); %把导向矢量作为权向量的限制向量 C*W = 1 Wc = C'*inv(C*C')*[1]; %定义上半支输入信号的权向量 %% 构造阻塞矩阵B 分别使用1、3、7个正交向量来构造 B1 = [-1,exp(-j*pi*sin(theta(1))),0,0,0,0,0,0]; % 1个正交向量构造阻塞矩阵 B2 = [-1,exp(-j*pi*sin(theta(1))),0,0,0,0,0,0; % 3个正交向量构造阻塞矩阵 0,-1,exp(-j*pi*sin(theta(1))),0,0,0,0,0; 0,0,-1,exp(-j*pi*sin(theta(1))),0,0,0,0;]; B3 = [-1,exp(-j*pi*sin(theta(1))),0,0,0,0,0,0; % 7个正交向量构造阻塞矩阵 0,-1,exp(-j*pi*sin(theta(1))),0,0,0,0,0; 0,0,-1,exp(-j*pi*sin(theta(1))),0,0,0,0; 0,0,0,-1,exp(-j*pi*sin(theta(1))),0,0,0; 0,0,0,0,-1,exp(-j*pi*sin(theta(1))),0,0; 0,0,0,0,0,-1,exp(-j*pi*sin(theta(1))),0; 0,0,0,0,0,0,-1,exp(-j*pi*sin(theta(1)))]; for i=1:1:M-1 alpha(:,i) = [1;exp(j*(2*pi*1*(0.5*sin(theta(1))+i/M)));exp(j*(2*pi*2*(0.5*sin(theta(1))+i/M)));exp(j*(2*pi*3*(0.5*sin(theta(1))+i/M)));exp(j*(2*pi*4*(0.5*sin(theta(1))+i/M)));exp(j*(2*pi*5*(0.5*sin(theta(1))+i/M)));exp(j*(2*pi*6*(0.5*sin(theta(1))+i/M)));exp(j*(2*pi*7*(0.5*sin(theta(1))+i/M)))]; end B0 = alpha'; % 另外一种方法构造的7*8阻塞矩阵 B = B0; %% 直接求出最优解 R = conj(y)*y.'/N; %计算自相关矩阵 Wa = (B*R*B')\B*R*Wc; %计算Wa W_GSC = Wc-B'*Wa; %计算GSC最优解 X = inv(C*inv(R)*C'); W_mvb = R\C'*X; %计算最小方差波束形成最优解 %% 自适应迭代 [L M] = size(B); %获取阻塞矩阵行数 c = 10000; %定义迭代次数 mu = 0.001; %定义迭代步长 Wa = zeros(L,c); %初始化迭代权值 for i=1:1:c yb = B*y(:,i); Za(i) = Wa(:,i).'*yb; Zc(i) = Wc.'*y(:,i); Z(i) = Zc(i)-Za(i); Wa(:,i+1) = Wa(:,i)+mu*Z(i)*conj(yb); end W_GSC_LMS = Wc-B'*Wa(:,end); %求最终权值 %% 画出阵列方向图 sita = [-90:0.5:90]*pi/180; %定义方向图角度自变量 L_sita = length(sita); %定义方向图角度自变量 for i =1:L_sita %遍历-90到90度，求出阵列方向增益 V(:,i) = exp(-j*pi*d'*sin(sita(i))); P(i) = V(:,i).'*W_GSC; end P0=abs(P); P0=P0./max(P0); %归一化信号功率增益 PdB = 20*log10(P0); figure(1); polar(sita,PdB+60); title('阵列方向图');grid on hold on degree=theta.'*ones(1,10); polar(degree(1,:),linspace(0,60,10),'r--'); polar(degree(2,:),linspace(0,60,10),'r--'); polar(degree(3,:),linspace(0,60,10),'r--'); polar(degree(4,:),linspace(0,60,10),'r--'); %% 计算输出信干噪比 % 输入信干噪比 -5.6dB signal_in = A(:,1)*S(1,:); %八根天线上信号的输入 signal_out = W_GSC.'*signal_in; %通过阵列后信号的输出 power_signal = mean(abs(signal_out).^2); %输出信号功率 all_out = W_GSC.'*y; %输出的全部信号 power_all = mean(abs(all_out).^2); %输出的全部信号功率 SINR = 10*log10(power_signal/(power_all-power_signal)) %计算输出信干噪比