untitled.asv

lms算法的自己编的程序 包括各种影响因素的分析等等 包括运行结果分析图

chanl_adp_model_define

step=0.003;
s_n=signal+noise;
expect=signal;
[h_all,y,erro]=chanl_adp_lms(s_n,expect(order_h+1:n_total),1,step);


figure(1)
plot(1:n_total-order_h,signal(order_h+1:n_total),'r')
legend('desired signal ');


figure(2)
plot(1:n_total,s_n(1:n_total),'g');
legend('the noise + the signal ');


%figure(3)
%plot(1:n_total-order_h,y,'b')
%legend('step=0.003 时的自适应滤波输出： ');

figure(6)

for i=1:4
    exp=3.^(i-1);
    [h_all,y,erro]=chanl_adp_lms(s_n,expect(order_h+1:n_total),exp,step);
    cost=erro.^2;
      for  j=1:length(erro)
           cost0(i,j)=mean(cost(1:j));
      end
       subplot(2,2,i),plot(1:n_total-order_h,y,'b');
end

figure(4) 
plot(1:length(erro),cost0(1,:),'r',1:length(erro),cost0(2,:),'b',1:length(erro),cost0(3,:),'g',1:length(erro),cost0(4,:),'k');

legend('\itmse offset=1','\itmse offset=3','\itmse offset=9','\itmse offset=27');
hold on

clear cost
clear i 
clear j


%-----------------------------------------------------------------
figure(3)

for i=1:4
   
    step=0.001*i;
    exp=1;
    [h_all,y,erro]=chanl_adp_lms(s_n,expect(order_h+1:n_total),exp,step);
    cost=erro.^2;
      for  j=1:length(erro)
           cost0(i,j)=mean(cost(1:j));
      end
       subplot(2,2,i),plot(1:n_total-order_h,y,'b');
end

figure(5) 
plot(1:length(erro),cost0(1,:),'r',1:length(erro),cost0(2,:),'b',1:length(erro),cost0(3,:),'g',1:length(erro),cost0(4,:),'k');

legend('\itstep=0.001','\itstep=0.002','\itstep=0.003','\itstep=0.004');
hold on
clear cost
clear i 
clear j