实现ASK调制解调,FSK调制解调,PSK调制解调,QAM调制解调的误码率仿真-源码

close all clear all clc % identifying Simulation parameters: n=1e6; % Number of bits/SNR=1 million bits SNR=0:2:30; % Signal to noise ratio range in dB with 2Db steps snr=10.^(SNR/10); % Signal to noise ratio %initializing empty matices to be filled with the sequences %OOK=3; OOK=zeros(1,16); PRK=zeros(1,16); FSK=zeros(1,16); OOK_mat=zeros(1,16); PRK_mat=zeros(1,16); FSK_mat=zeros(1,16); QAM_mat=zeros(1,16); % Transmitter side variables bits=randi([0,1],1,n); %Generate random binary data vector(1 x n) %we will work on ASK PSK FSK modulation according to the specified cases in the manual %ASK special case: OOK_Modulation(0,1) signal_OOK=bits; %No change in the bits will be required %PSK special case: PRK modulation(-1,1) signal_PRK=(bits.*2)-1; %represent the 1 by 1 and the 0 bit by -1 using provided formula %FSK special case: Orthogonal-FSK modulation ones=find(bits); %find all elements that are equal to one in the bits sequence zeros=find(~bits); %find all elements that are equal to zero in the bits sequence signal_FSK(ones)=i; %replace all the ones with i which is an imaginary number signal_FSK(zeros)=1; %replace all the zeros with 1 % modulating using matlab built in function OOK_mat_mod=genqammod(bits,[0 1]); %OOK modulation using built in functions PRK_mat_mod=pskmod(bits,2); %PRK modulation using built in functions FSK_mat_mod=genqammod(bits,[1 1i]); %FSK modulation using built in functions rbdv=randi([0 16-1],1,n); %Generate random data vector(1 x n) QAM_mat_mod=qammod(rbdv,16); %QAM modulation using built in functions %Applying noise to the bit stream with different values of SNR,receiving %the bits stream,recovering the original bits by demodulating it %and calculating the BER and Pe for each type of modulated signal for i=1:16 Ptx_OOK=mean(signal_OOK.^2) noise=(sqrt(mean(abs(signal_OOK).^2)/(2*snr(i))))*(randn(1,n)+(1j*randn(1,n)));%applying the noise on OOK sequence OOK_sequence=signal_OOK+noise; Ptx_PRK=mean(signal_PRK.^2) noise=(sqrt(mean(abs(signal_PRK).^2)/(2*snr(i))))*(randn(1,n)+(1j*randn(1,n)));%applying the noise on PRK sequence PRK_sequence=signal_PRK+noise; Ptx_FSK=mean(signal_FSK.^2) noise=(sqrt(mean(abs(signal_FSK).^2)/(2*snr(i))))*(randn(1,n)+(1j*randn(1,n)));%applying the noise on FSK sequence FSK_sequence=signal_FSK+noise; %aplying noise on signals after modulation using functions noise=(sqrt(mean(abs(OOK_mat_mod).^2)/(2*snr(i))))*(randn(1,n)+(1j*randn(1,n)));%applying the noise on OOK_tb sequence OOK_sequence_mat=OOK_mat_mod+noise; noise=(sqrt(mean(abs(PRK_mat_mod).^2)/(2*snr(i))))*(randn(1,n)+(1j*randn(1,n)));%applying the noise on PRK_tb sequence PRK_sequence_mat=PRK_mat_mod+noise; noise=(sqrt(mean(abs(FSK_mat_mod).^2)/(2*snr(i))))*(randn(1,n)+(1j*randn(1,n)));%applying the noise on FSK_tb sequence FSK_sequence_mat=FSK_mat_mod+noise; noise=(sqrt(mean(abs(QAM_mat_mod).^2)/(2*snr(i))))*(randn(1,n)+(1j*randn(1,n)));%applying the noise on QAM_tb sequence QAM_sequence_mat=QAM_mat_mod+noise; %Decide whether the Rx_sequence is �1� or �0� by comparing each bit with a threshold OOK_demod=(real(OOK_sequence) >=0.5);%0.5 for OOK PRK_demod=(real(PRK_sequence) >=0);%0 for PRK FSK_demod=(real(FSK_sequence)<imag(FSK_sequence));%in FSK if real > imaginary component then it equal zero else equals one %demodulation using built in functions OOK_mat_demod = genqamdemod(OOK_sequence_mat,[0 1]); PRK_mat_demod = pskdemod(PRK_sequence_mat,2); FSK_mat_demod = genqamdemod(FSK_sequence_mat,[1 1i]); rbdv_mat_demod = qamdemod(QAM_sequence_mat,16); %Calculate #of errors (BER) %[NUMBER,RATIO] = BITERR(X,Y) [~,OOK(i)]=biterr(bits,OOK_demod); [~,PRK(i)]=biterr(bits,PRK_demod); [~,FSK(i)]=biterr(bits,FSK_demod); %for modulated using matlab [~,OOK_mat(i)] = biterr(bits,OOK_mat_demod); [~,PRK_mat(i)] = biterr(bits,PRK_mat_demod); [~,FSK_mat(i)] = biterr(bits,FSK_mat_demod); [~,QAM_mat(i)] = symerr(rbdv,rbdv_mat_demod); end % constellation n=1e6; snrq=15; refC = qammod(0:15,16); constDiagram = comm.ConstellationDiagram('ReferenceConstellation',refC , 'XLimits',[-4 4],'YLimits',[-4 4]); data = randi([0 15],n,1); sym = qammod(data,16); rcv = awgn(sym,snrq); step(constDiagram,rcv) % Ploting all in one curve figure semilogy(SNR,OOK,'r-*','LineWidth',2) hold on; semilogy(SNR,OOK_mat,'m--o','LineWidth',2) hold on; semilogy(SNR,PRK,'g-*','LineWidth',2) hold on; semilogy(SNR,PRK_mat,'k--+','LineWidth',2) hold on; semilogy(SNR,FSK,'b-*','LineWidth',2) hold on; semilogy(SNR,FSK_mat,'y--s','LineWidth',2) hold on; semilogy(SNR,QAM_mat,'c-p','LineWidth',2) title('BER vs. SNR ') ylabel('BER') xlabel('SNR') legend('OOK','OOK-Function','PRK','PRK-Function','FSK','FSK-Function','QAM-Function') grid on;