Channel Estimation in OFDM using Neural Networks.

I am trying to implement the following code, but i get lots of fluctuations in MSE vs SNR of Neural networks. Could anyone please help me in getting a smoother response.

The code is give below:

clc clear all

nCP = 8; %round(Tcp/Ts); nFFT = 32; NT = nFFT + nCP; F = dftmtx(nFFT)/sqrt(nFFT); MC = 2; EsNodB = 0:1:31; snr = 10.^(EsNodB/10); beta = 17/9; M = 16; modObj = modem.qammod(M); demodObj = modem.qamdemod(M); L = 5; for ii = 1:length(EsNodB) disp('EsN0dB is :'); disp(EsNodB(ii)); tic; ChMSE_LS = 0; error=zeros(1,32); err2=[] err1=[] for mc = 1:MC % Random channel taps g = randn(L,1)+1i*randn(L,1); g = g/norm(g); H = fft(g,nFFT); % generation of symbol X = randi([0 M-1],nFFT,1); %BPSK symbols XD = modulate(modObj,X)/sqrt(10); % normalizing symbol power x = F'*XD; %loading xout = [x(nFFT-nCP+1:nFFT);x]; % channel convolution and AWGN y = conv(xout,g); nt =randn(nFFT+nCP+L-1,1) + 1i*randn(nFFT+nCP+L-1,1); No = 10^(-EsNodB(ii)/10); y = y + sqrt(No/2)*nt; % Receiver processing y = y(nCP+1:NT); Y = F*y; % frequency doimain LS channel estimation HhatLS = Y./XD; ChMSE_LS = ChMSE_LS + ((H -HhatLS)'*(H-HhatLS))/nFFT; %for l=1:32 x1=abs(x); x2=x1'; y1=abs(Y); y2=y1'; net=newff(minmax(y2), [4,1], {'logsig','purelin', 'trainlm'}); net=init(net) net.trainparam.epochs=600; %No. of iterations net.trainparam.lr=0.01; [net,tr]=train(net,y2,x2) x_net = net(y2); error=(x2-x_net); %err1=sum(err) %err2(l)=err1/MC; dbstop if error % end end %inp=rand(1,4) %v=sim(net,inp) %plot(v); %l= v-inp;

ChEstLS(ii) = ChMSE_LS/MC; %error1(ii)=err/MC; err=error/MC; end toc figure, plot(err,'g'); legend('Error'); figure, % Channel estimation semilogy(EsNodB,abs(err),'b','LineWidth',2); hold on semilogy(EsNodB,ChEstLS,'r','LineWidth',2); hold on;grid on;xlabel('EsNodB'); ylabel('Channel MSE');

legend('NN', 'LS');

[Abdelwahab Afifi]

could you uplaod the code first in clear format?

[Joswa]

can you provide the dataset for this code??

[Sajid Sarwar]

Where to get dataset for this simulation?