MATLAB实现基于OFDM 的802.11a系统及仿真.zip

%IEEE 802.11a系统的仿真 clear all %%%参数设置部分%%% Nsp = 52; %系统子载波数，（不包括直流载波） Nfft = 64; %FFT长度 Ncp = 16; %循环前缀长度 Ns = Nfft+Ncp; %一个完整OFDM符号长度 noc = 53; %包含直流载波的总的子载波数 Nd = 6; %每帧包含的OFDM符号数（不包括训练符号） M1 = 4; %QPSK调制 M2 = 16; %16-QAM调制 sr = 250000; %OFDM符号速率 EbN0 = 0:2:30; %归一化信噪比 Nfrm = 10000; %每种信噪比下的仿真帧数 ts = 1/sr/Ns; %OFDM符号抽样时间间隔 t = 0:ts:(Ns*(Nd+1)*Nfrm-1)*ts; %抽样时刻 fd = 100; %最大多普勒频移 h = rayleigh(fd,t); %生成单径Rayleigh衰落信道 %训练符号频域数据，采用802.11a中的长训练符号数据 Preamble=[1 1 -1 -1 1 1 -1 1 -1 1 1 1 1 1 1 -1 -1 1 1 -1 1 -1 1 1 1 1 ... 1 -1 -1 1 1 -1 1 -1 1 -1 -1 -1 -1 -1 1 1 -1 -1 1 -1 1 -1 1 1 1 1]; Preamble1=zeros(1,Nfft); Preamble1(2:27)=Preamble(27:end); %训练符号重排后的数据??? Preamble1(39:end)=Preamble(1:26); preamble1=ifft(Preamble1); %训练符合时域数据??? preamble1=[preamble1(Nfft-Ncp+1:end) preamble1]; %加入循环前缀 %%%仿真循环%%% for ii=1:length(EbN0) %%%发射机部分%%% msg1=randsrc(Nsp,Nd*Nfrm,[0:M1-1]); %QPSK信息数据?? msg2=randsrc(Nsp,Nd*Nfrm,[0:M2-1]); %16-QAM信息数据?? data1=pskmod(msg1,M1,pi/4); %QPSK调制?? data2=qammod(msg2,M2)/sqrt(10); %16-QAM调制并归一化 data3=zeros(Nfft,Nd*Nfrm); %根据FFT要求，对数据重排?????????? data4=zeros(Nfft,Nd*Nfrm); data3(2:27,:)=data1(27:end,:); %重排后的数据 data3(39:end,:)=data1(1:26,:); data4(2:27,:)=data2(27:end,:); %重排后的数据 data4(39:end,:)=data2(1:26,:); clear data1 data2; %清除不需要的临时变量? data3=ifft(data3); %IFFT变换 data4=ifft(data4); data3=[data3(Nfft-Ncp+1:end,:);data3]; %加入循环前缀 data4=[data4(Nfft-Ncp+1:end,:);data4]; spow1=norm(data3,'fro').^2/(Nsp*Nd*Nfrm); %计算数据符号能量??????????? spow2=norm(data4,'fro').^2/(Nsp*Nd*Nfrm); data5=zeros(Ns,(Nd+1)*Nfrm); %加入训练符号??? data6=data5; for indx=1:Nfrm data5(:,(indx-1)*(Nd+1)+1)=preamble1.'; data5(:,(indx-1)*(Nd+1)+2:indx*(Nd+1))=data3(:,(indx-1)*Nd+1:indx*Nd); data6(:,(indx-1)*(Nd+1)+1)=preamble1.'; data6(:,(indx-1)*(Nd+1)+2:indx*(Nd+1))=data4(:,(indx-1)*Nd+1:indx*Nd); end clear data3 data4 data5=reshape(data5,1,Ns*(Nd+1)*Nfrm); %并串变换 data6=reshape(data6,1,Ns*(Nd+1)*Nfrm); sigma1=sqrt(1/2*spow1/log2(M1)*10.^(-EbN0(ii)/10)); %根据EbN0计算噪声标准差???? sigma2=sqrt(1/2*spow2/log2(M2)*10.^(-EbN0(ii)/10)); for indx=1:Nfrm dd1=data5((indx-1)*Ns*(Nd+1)+1:indx*Ns*(Nd+1)); %当前帧的发射数据????? dd2=data6((indx-1)*Ns*(Nd+1)+1:indx*Ns*(Nd+1)); hh=h((indx-1)*Ns*(Nd+1)+1:indx*Ns*(Nd+1)); %当前帧对应的信道参数?? %信号通过单径Rayleigh衰落信道，并加入高斯白噪声???????????????? r1=hh.*dd1+sigma1*(randn(1,length(dd1))+j*randn(1,length(dd1))); r2=hh.*dd2+sigma2*(randn(1,length(dd2))+j*randn(1,length(dd2))); r1=reshape(r1,Ns,Nd+1); %串并变换 r2=reshape(r2,Ns,Nd+1); r1=r1(Ncp+1:end,:); %移除循环前缀 r2=r2(Ncp+1:end,:); %%%理想信道估计%%% hh=reshape(hh,Ns,Nd+1); %信道参数数据重排????????? hh=hh(Ncp+1:end,:); x1=r1(:,2:end)./hh(:,2:end); %理想信道估计结果 x2=r2(:,2:end)./hh(:,2:end); x1=fft(x1); %FFT运算?? x2=fft(x2); x1=[x1(39:end,:);x1(2:27,:)]; %数据重排??????? x2=[x2(39:end,:);x2(2:27,:)]; x1=pskdemod(x1,M1,pi/4); %数据解调????? x2=qamdemod(x2*sqrt(10),M2); [neb1(indx),temp]=biterr(x1,msg1(:,(indx-1)*Nd+1:indx*Nd),log2(M1));%统计一帧中的错误比?????? [neb2(indx),temp]=biterr(x2,msg2(:,(indx-1)*Nd+1:indx*Nd),log2(M2)); %%%根据训练符号进行的信道估计%%% R1=fft(r1); %fft运算? R2=fft(r2); R1=[R1(39:end,:);R1(2:27,:)]; %数据重排?????? R2=[R2(39:end,:);R2(2:27,:)]; HH1=(Preamble.')./R1(:,1); %信道估计 ??? HH2=(Preamble.')./R2(:,1); HH1=HH1*ones(1,Nd); %根据信道估计结果进行信道补偿?? HH2=HH2*ones(1,Nd); x3=R1(:,2:end).*HH1; x4=R2(:,2:end).*HH2; x3=pskdemod(x3,M1,pi/4); %数据解调???? x4=qamdemod(x4.*sqrt(10),M2); [neb3(indx),temp]=biterr(x3,msg1(:,(indx-1)*Nd+1:indx*Nd),log2(M1)); %统计一帧中的错误比?????? [neb4(indx),temp]=biterr(x4,msg2(:,(indx-1)*Nd+1:indx*Nd),log2(M2)); end ber1(ii)=sum(neb1)/(Nsp*log2(M1)*Nd*Nfrm); %理想信道估计的误比特率????? ber2(ii)=sum(neb2)/(Nsp*log2(M2)*Nd*Nfrm); ber3(ii)=sum(neb3)/(Nsp*log2(M1)*Nd*Nfrm); %根据训练符号信道估计的误比特率??????? ber4(ii)=sum(neb4)/(Nsp*log2(M2)*Nd*Nfrm); end semilogy(EbN0,ber1,'-ro',EbN0,ber3,'-rv',EbN0,ber2,'-g*',EbN0,ber4,'-gp') grid on title("OFDM系统误比特率性能") legend("QPSK理想信道估计","QPSK训练符号信道估计", "16-QAM理想信道估计","16-QAM训练符号信道估计") xlabel("信噪比（EbN0）") ylabel("误比特率")