the following code doesnt give me the ber vs ebno i need which is a logarthmic curve..all i get is a noise signal..can someone tell me whats wrong with it

Question

v@ishu 2015-3-3

0
链接

此问题的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/181305-the-following-code-doesnt-give-me-the-ber-vs-ebno-i-need-which-is-a-logarthmic-curve-all-i-get-is-a

编辑： v@ishu 2015-3-3

clc; clear all; close all; nbitpersym = 256; % number of bits per OFDM symbol nsym = 1000; % number of symbols len_fft =256; % fft size sub_car =256; % number of data subcarriers EbNo = 0:2:30; snr=EbNo+(10*log(524/512));

M = modem.pskmod(4); % modulation object

% Generating data t_data=randint(nbitpersym*nsym,1);

% serial to parallel conversion par_data = reshape(t_data,nsym,sub_car); % modulating data mod_data = modulate(M,par_data);

% making X[k], -X[k] subcarriers par_data_self=zeros(nsym,sub_car*2); for i=1:sub_car, par_data_self(:,2*i-1)=mod_data(:,i); par_data_self(:,2*i)=(exp(-j*pi/2))*((mod_data(:,i))); end

% IFFT transform for OFDM transmission IFFT_data = ifft(par_data_self);

% addition cyclic prefix cylic_add_data = [IFFT_data(:,[501:512]) IFFT_data];

% parallel to serial coversion ser_data = reshape(cylic_add_data,524*nsym,1); K=length(ser_data); ep=0:9.5420e-007:0.5; k=1:(length(ep));

%phase noise addition d=(ep(k))./(K*(0:K-1)); d1=reshape(d,524000,1); noised_data = ser_data.*exp(j*pi*d1); no_of_error=[]; ratio=[]; for ii=1:length(snr)

% chan_awgn = awgn(ser_data,snr(ii));% awgn addition

ser_to_para = reshape(noised_data,524,1000).'; % serial to parallel coversion

cyclic_pre_rem = ser_to_para(:,[13:524]); %cyclic prefix removal

FFT_recdata = fft(cyclic_pre_rem);% freq domain transform

% making self canceled data by 0.5*(X[k]-(-X[k]) rx=zeros(nsym,sub_car); for i=1:sub_car, rx(:,i)=0.5*(FFT_recdata(:,2*i-1)-(exp(-j*pi/2)*FFT_recdata(:,2*i))); end

ser_data_rx = reshape(rx.',nbitpersym*nsym,1); % serial coversion

z=modem.pskdemod(4); %demodulation object

demod_Data = demodulate(z,ser_data_rx); %demodulating the data [no_of_error(ii),ratio(ii)]=biterr(t_data,demod_Data) end % figure(1); % stem(t_data); % hold on; % stem(demod_Data,'*'); % title('plural weighted data conversion scheme t_data and received data compare'); figure(2); semilogy(EbNo,ratio); xlabel('EbNo'); ylabel('BER') title('Bit error rate of plural weighted data conversion scheme');