trial code 2.rar_adaboost matlab_nines2q

%% OFDM system with channel estimation: clear all; close all; clc; global W %% define parameters m=input( ' how many OFDM symbols to be simulated m = ' ) ; N=input( ' length of OFDM symbols N = ' ) ; M=input( ' Constellation Order M = ' ) ; pilotFrequency=input('Pilot Frequency = ' ) ; E=input('Pilot Energy = '); Ncp=input( ' Cyclic Prefix length = ' ) ; L=input(' channel Length (number of taps) = '); typ=input('Constellation Family Type is [1 for M-QAM, 2 for M-PSK]:- '); %% ---------------------------------------------------------------------- %% define the modems Tx/Rx switch typ case 1 Tx=modem.qammod('M',M); Rx=modem.qamdemod ('M',M); case 2 Tx=modem.pskmod('M',M); Rx=modem.pskdemod ('M',M); otherwise error('Error, Constellation Family not Defined'); end %% data generation D=randi ([0 M-1],m,N); %% mapping (baseband modulation ) D_Mod=modulate(Tx,D); %% serial to parallel D_Mod_serial=D_Mod.'; %% specify Pilot & Date Locations PLoc = 1:pilotFrequency:N; % location of pilots DLoc = setxor(1:N,PLoc); % location of data %% Pilot Insertion D_Mod_serial(PLoc,:)=E*D_Mod_serial(PLoc,:); figure; imagesc(abs(D_Mod_serial )) %% inverse discret Fourier transform (IFFT) % Amplitude Modulation d_ifft=ifft(D_Mod_serial); %% parallel to serail d_ifft_parallel=d_ifft.'; %% Adding Cyclic Prefix CP_part=d_ifft_parallel(:,end-Ncp+1:end); % this is the Cyclic Prefix part to be appended. ofdm_cp=[CP_part d_ifft_parallel]; %% generating random channel % Rayleigh Channel Model % Input : L : # of channel realization % Output: H : Channel vector %MIMO-OFDM Wireless Communications with MATLAB�� Yong Soo Cho, Jaekwon Kim, Won Young Yang and Chung G. Kang %2010 John Wiley & Sons (Asia) Pte Ltd h = (randn(1,L)+j*randn(1,L))/sqrt(2); h = h./norm(h); % normalization H = fft(h,N); % Frequency-Domain Channel d_channelled = filter(h,1,ofdm_cp.').'; % channel effect channel_length = length(h); % True channel and its time-domain length H_power_dB = 10*log10(abs(H.*conj(H))); % True channel power in dB %% add noise count=0; snr_vector=0:4:40; for snr=snr_vector SNR = snr + 10*log10(log2(M)); count=count+1 ; disp(['step: ',num2str(count),' of: ',num2str(length(snr_vector))]) ofdm_noisy_NoCH=awgn(ofdm_cp,SNR,'measured' ) ; ofdm_noisy_with_chann=awgn(d_channelled,SNR,'measured' ) ; %% receiver %Remove Cyclic Prefix ofdm_cp_removed_NoCH=ofdm_noisy_NoCH(:,Ncp+1:N+Ncp); ofdm_cp_removed_with_chann=ofdm_noisy_with_chann(:,Ncp+1:N+Ncp); % serial to parallel ofdm_parallel_NoCH=ofdm_cp_removed_NoCH.'; ofdm_parallel_chann=ofdm_cp_removed_with_chann.'; %% Discret Fourier transform (FFT) % Amplitude Demodulation d_parallel_fft_NoCH=fft(ofdm_parallel_NoCH) ; d_parallel_fft_channel=fft(ofdm_parallel_chann) ; %% channel estimation % Extracting received pilots TxP = D_Mod_serial(PLoc,:); % trnasmitted pilots RxP = d_parallel_fft_channel(PLoc,:); % received pilots % Least-Square Estimation Hpilot_LS= RxP./TxP; % LS channel estimation % MMSE Estimation:- for r=1:m H_MMSE(:,r) = MMSE(RxP(:,r),TxP(:,r),N,pilotFrequency,h,SNR); end % Interpolation p--->N for q=1:m HData_LS(:,q) = interpolate(Hpilot_LS(:,q).',PLoc,N,'spline'); % Linear/Spline interpolation end %% parallel to serial HData_LS_parallel1=HData_LS.'; HData_MMSE_parallel1=H_MMSE.'; %% demapping d_received_NoCH=demodulate(Rx,(d_parallel_fft_NoCH.')) ; % No Channel d_received_chann_LS=demodulate(Rx,(d_parallel_fft_channel.')./HData_LS_parallel1) ; % LS channel estimation d_received_chann_MMSE=demodulate(Rx,(d_parallel_fft_channel.')./(HData_MMSE_parallel1)) ; % MMSE channel estimation %% Removing Pilots from received data and original data D_no_pilots=D(:,DLoc); % removing pilots from D Rec_d_NoCH=d_received_NoCH(:,DLoc); % removing pilots from d_received_NoCH Rec_d_LS=d_received_chann_LS(:,DLoc); % removing pilots from d_received_chann_LS Rec_d_MMSE=d_received_chann_MMSE(:,DLoc); % removing pilots from d_received_chann_MMSE %% Calculating BER [~,r_NoCH(count)]=symerr(D_no_pilots,Rec_d_NoCH) ; [~,r_LS(count)]=symerr(D_no_pilots,Rec_d_LS) ; [~,r_MMSE(count)]=symerr(D_no_pilots,Rec_d_MMSE) ; end figure; semilogy(snr_vector,r_NoCH,'-+');hold on semilogy(snr_vector,r_LS,'-o'); semilogy(snr_vector,r_MMSE,'-s'); legend('orig. No Channel','LS CE','MMSE CE'); grid ; hold off; H_power_esti_dB_LS = 10*log10(abs(HData_LS_parallel1.*conj(HData_LS_parallel1))); % Estimated channel power in dB H_power_esti_dB_MMSE = 10*log10(abs(HData_MMSE_parallel1.*conj(HData_MMSE_parallel1))); % Estimated channel power in dB figure;hold on; plot(H_power_dB(1:8:end),'+k','LineWidth',3); plot(H_power_esti_dB_LS(1,(1:8:end)),'or','LineWidth',3); plot(H_power_esti_dB_MMSE(1,(1:8:end)),'Sb','LineWidth',1); title('ACTUAL AND ESTIMATED CHANNELS'); xlabel('Time in samples'); ylabel('Magnitude of coefficients'); legend('Actual','estimated LSE','estimated MMSE')