I am doing a MATLAB homework about adaptive equalizer.

I have to insert some code in bold sentence and make a LMS algorithm.

Final result should be plotted like a third picture.

first and second picture may be a hint. please help me.

% % Test 1 % Design of Adaptive Equalizer: Impact of bandwdith on MSE performance % clear all; close all; clc; % Parameter Generation rp.Nruns = 200; % No. of runs rp.var_v = 0.001; % Noise Variance rp.p = 11; % No. of taps in equlizer rp.tau = 7; % Delay rp.Ndata = 500; % No. of input data x_n rp.mu = 0.075; % step-size for LMS % Ch_W = [2.9 3.1 3.3 3.5]; % Channel bandwidth Z = zeros(rp.Ndata - rp.tau,length(Ch_W)); % For mean-square error % tstart = tic; % for i=1:length(Ch_W) rp.T = Ch_W(i); % seed = 0:(rp.Nruns-1); % Seed Generation h = [0 0.5*(1+cos((2*pi/rp.T)*((1:3) - 2)))]; % Channel Generation % Npred = rp.Ndata - rp.tau; % No. of iterations in LMS E = zeros(Npred, rp.Nruns); % Error matrix (Npred x rp.Nruns) in LMS % for run_iter = 1:rp.Nruns, % rand( 'seed', seed(run_iter)); % For input data randn('seed', seed(run_iter)); % For AWGN % a = ; % Generation of input data (+1 or -1) Xi = conv(a, h); % Channel output Xi = Xi(1:length(a)); % Xi = Xi + ; % Addition of AWGN % Xi = [zeros(rp.p - rp.tau, 1); Xi]; % Delay adjustment % % LMS algorithm

% % % E(:, run_iter) = e; % error at each run end; % for run_iter % Z(:,i)=sum((abs(E).^2),2)/rp.Nruns; % Mean-square error end % for Ch_W(i) % time_elapsed = toc(tstart) % semilogy(Z(:,4),'b'); hold on; grid on; semilogy(Z(:,3),'r'); semilogy(Z(:,2),'k'); semilogy(Z(:,1),'m'); % xlabel('Number of iterations, n'); ylabel('Ensemble-average-square error'); legend('W = 3.5','W = 3.3','W = 3.1','W = 2.9'); %

Adaptive Equalizer Delay Adaptive Random-noise Channel generator (1) xn transversal equalizer (n) Random-noise generator (2) Figure 5.19 Block diagram of adaptive equalizer experiment (n)