非线性BD-GMD均分功率用户排序误码率matlab仿真,含仿真操作录像

clc; clear; close all; warning off; addpath(genpath(pwd)); factor=0; %设置仿真参数 Nt =16; % Tx 天线 Nr =16; % Rx 天线 %UserAntennaNum的累加和为H的行数 UserNum=4; UserAntennaNum_old=[ 4 4 4 4];%对比不同天线数目下，BER性能 stream_num_old=[ 4 4 4 4]; %stream_num=[3 3 3;2 2 5;2 3 4;3 4 2;2 4 3; 5 2 2]; %stream_num=[3 3 4;2 2 6;2 3 4;4 4 2;2 4 4; 6 2 2]; [row_test col_test]=size(UserAntennaNum_old); SymPerFrame = 1; Frame_num = 50; % 仿真中帧的数目 M =16;% QAM调制 P_all=5; % the sum power of all stream is P_all %PowerScale= sqrt(2*(M-1)/3); ModPeriod = 2*sqrt(M) ;% mod操作数，用于THP预编码 % 设置信噪比范围 SNR = [-5,-3:3:24]; %初始化误比特率 Len = length(SNR); ber_bdgmd_noorder=zeros(row_test,Len); ber_bdgmd_optimalorder=zeros(row_test,Len); %PowerScale = 2*M/3;% 功率归一化 PowerScale_linear_GMD=2*M/3; %由于星座点映射之后的功率时2(M-1)/3,再经过mod之后放大M/(M-1),所以mod后功率为PowerScale，而每个流上的最终功率为 %SigPowerDbw_of_each_stream，所以beta为功率控制系数，可为系数也可为对角矩阵 %-----------------循环过程开始----------------- for index = 1:Len;%SNR取值变化循环 %--------------------- to send the program's progress------------------- fprintf('%s %d %s \n', 'The progress of this program is ',SNR(index)/max(SNR)*100,'%'); snr = 10.^(SNR(index)/10);% 将SNR转为线性域的变化范围 % noise_power = Nt/snr;% 噪声功率 %-----------------帧的变化范围，在每一个帧内信道保持不变----------------- for cnt = 1:Frame_num H1 = sqrt(1/2)*(randn(UserAntennaNum_old(1),Nt)+j*randn(UserAntennaNum_old(1),Nt));%随机生成信道矩阵 H2 = sqrt(1/2)*(randn(UserAntennaNum_old(2),Nt)+j*randn(UserAntennaNum_old(2),Nt));%随机生成信道矩阵 H3 = sqrt(1/2)*(randn(UserAntennaNum_old(3),Nt)+j*randn(UserAntennaNum_old(3),Nt));%随机生成信道矩阵 H4 = sqrt(1/2)*(randn(UserAntennaNum_old(3),Nt)+j*randn(UserAntennaNum_old(3),Nt));%随机生成信道矩阵 %% no user order 均分功率 %用户排序 % [H UserAntennaNum stream_num]=random_order(H1,H2,H3,H4,UserAntennaNum_old,stream_num_old); %-----------------每一帧中包含2*100个符号，循环从1到100----------------- for sym_index = 1:SymPerFrame; H_est=H; %完美信道信息 for n=1:row_test %n代表同时配置的几种天线接收方案 %--------------------产生信号源------------------------ Source = randint(sum(stream_num(n,:)),1,M);%生成2*1的符号，符号取值范围是0~15 Sym = qammod(Source,M,0,'gray'); %进行QAM调制和功率归一化 SigPower_of_each_stream=P_all/sum(stream_num(n,:));%按流均分功率时，将要发射时stream功率 SigPowerDbw_of_each_stream=10*log10(SigPower_of_each_stream); %----------------非线性BD-GMD预编码---------------------------- % 功率分配 beta_nonlinear_GMD=sqrt(SigPower_of_each_stream/PowerScale_linear_GMD); % 预编码 [d_thp,L,Q,G,PPP]= GMD_encoder(Sym,H_est,ModPeriod,UserNum,UserAntennaNum(n,:),stream_num(n,:),beta_nonlinear_GMD); % @para Sym ,just is one MQAM symbol,it is a complex number. % @para H_est ,the radnom channel which has been added the % noise,here it is a 4*4 complex matrix. % beta_water_BD_GMD功率分配矩阵求倒数 %[ beta_water_BDGMD_mine_inverse nonzero_stream_index]=water_inverse(diag(beta_water_BDGMD_mine)); % MIMO channel HH=H*d_thp; y_thp = awgn(HH,SNR(index),SigPowerDbw_of_each_stream); % 注水定理只体现在噪声上 % Caculate channel capacity G=diag(1./diag(G)); % receiver end r_thp = G*Q'*y_thp/beta_nonlinear_GMD; Rec_thp = mod_thp(r_thp,ModPeriod);% mod % QAM demodulator % Rec_thp = Rec_thp_1;% Scale the symbol to QAM symbol Rec_Data_thp = qamdemod(Rec_thp,M,0,'gray'); % BER 只统计功率分配不为0 的流的误码率 [err ratio] = biterr(Rec_Data_thp,Source,log2(M));% ber ber_bdgmd_noorder(n,index) = ber_bdgmd_noorder(n,index) + err; % 计算总流数 %stream_num_BD_GMD_mine_total(n,index)=stream_num_BD_GMD_mine_total(n,index)+length(nonzero_stream_index); end end; % loop for num %% 功率分配 遍历排序 [H,UserAntennaNum,stream_num]=optimal_order(H1,H2,H3,H4,UserNum,UserAntennaNum_old,stream_num_old); for sym_index = 1:SymPerFrame; H_est=H; %完美信道信息 for n=1:row_test %n代表同时配置的几种天线接收方案 %--------------------产生信号源------------------------ Source = randint(sum(stream_num(n,:)),1,M);%生成2*1的符号，符号取值范围是0~15 Sym = qammod(Source,M,0,'gray'); %进行QAM调制和功率归一化 SigPower_of_each_stream=P_all/sum(stream_num(n,:));%按流均分功率时，将要发射时stream功率 SigPowerDbw_of_each_stream=10*log10(SigPower_of_each_stream); % 功率分配 beta_nonlinear_GMD=sqrt(SigPower_of_each_stream/PowerScale_linear_GMD); % 预编码 [d_thp,L,Q,G,PPP]=GMD_encoder(Sym,H_est,ModPeriod,UserNum,UserAntennaNum(n,:),stream_num(n,:),beta_nonlinear_GMD); % @para Sym ,just is one MQAM symbol,it is a complex number. % @para H_est ,the radnom channel which has been added the % noise,here it is a 4*4 complex matrix. % beta_water_BD_GMD功率分配矩阵求倒数 %[ beta_water_BDGMD_mine_inverse nonzero_stream_index]=water_inverse(diag(beta_water_BDGMD_mine)); % MIMO channel HH=H*d_thp; y_thp = awgn(HH,SNR(index),SigPowerDbw_of_each_stream); % 注水定理只体现在噪声上 %noise=y_thp-HH; %y_thp_wf=HH+diag(beta_water_BDGMD_mine_inverse)*noise; % Caculate channel capacity G=diag(1./diag(G)); % receiver end r_thp = G*Q'*y_thp/beta_nonlinear_GMD; Rec_thp = mod_thp(r_thp,ModPeriod);% mod % QAM demodulator % Rec_thp = Rec_thp_1;% Scale the symbol to QAM symbol Rec_Data_thp = qamdemod(Rec_thp,M,0,'gray'); % BER 只统计功率分配不为0 的流的误码率 [err ratio] = biterr(Rec_Data_thp,Source,log2(M));% ber ber_bdgmd_optimalorder(n,index) = ber_bdgmd_optimalorder(n,index) + err; %err_rem((cnt-1)*SymPerFrame+sym_index)=err; % 计算总流数 %stream_num_BD_GMD_mine_total(n,index)=stream_num_BD_GMD_mine_total(n,index)+length(nonzero_stream_index); end end end; % loop for iteration end % loop for snr %----------------------------循环结束---------------------------- stream_num_of_per_loop=sum(stream_num); ber_bdgmd_noorder=ber_bdgmd_noorder./(Frame_num*SymPerFrame*stream_num_of_per_loop*sqrt(M));%计算误比特率 ber_bdgmd_optimalorder=ber_bdgmd_optimalorder./(Frame_num*SymPerFrame*stream_num_of_per_loop*sqrt(M)); %[3 3 4;2 2 6;2 3 5;4 4 2;2 4 4; 6 2 2] %----------------保存当前workspace中所有变量值到filename.mat文件中--------- filename='Linear-BD-GMD-SVD-THP-16QAM-4user-uesrodrder-streamreduced-40000';%-----需要改，根据程序不同改名 nowtime=clock; filena