MATLAB实现正交频分多路复用技术

% =================OFDM仿真参数说明：================ % % 子载波数 carrier_count ---200 % 总符号数 symbol_count ---100 % IFFT长度 ifft_length ---512 % 循环前缀 CP_length ---512/4=128 % 循环后缀 CS_length ---20 % 升余弦窗系数 alpha ---7/32 % 调制方式 QAM16、QPSK 可选 % 多径幅度 mult_path_am ---[1 0.2 0.1] % 多径时延 mutt_path_time ---[0 20 50] % ====================仿真过程======================= % 产生0-1随机序列 => 串并转换 => 映射 => 取共轭、过采样 % => IFFT => 加循环前缀和后缀 => 加窗 => 并串转换 => % 多径信道 => 加AWGN => 串并转换 => 去前缀 => FFT => % 下采样 => 解映射 => 求误码率 % ================================================== clear all; close all; carrier_count = 200; % 子载波数 symbol_count = 100; ifft_length = 512; CP_length = 128; CS_length = 20; rate = []; SNR =20; bit_per_symbol = 4; % 调制方式决定 alpha = 1.5/32; % ================产生随机序列======================= bit_length = carrier_count*symbol_count*bit_per_symbol; bit_sequence = round(rand(1,bit_length))'; % 列向量 % ================子载波调制方式1======================== % 1-28置零 29-228有效 229-285置零 286-485共轭 486-512置零 carrier_position = 29:228; conj_position = 485:-1:286; bit_moded = qammod(bit_sequence,16,'InputType','bit'); figure('position',[0 0 400 400],'menubar','none'); scatter(real(bit_moded),imag(bit_moded)); title('调制后的散点图'); grid on; % ===================IFFT=========================== % =================串并转换========================== ifft_position = zeros(ifft_length,symbol_count); bit_moded = reshape(bit_moded,carrier_count,symbol_count); figure('position',[400 0 400 400],'menubar','none'); stem(abs(bit_moded(:,1))); grid on; ifft_position(carrier_position,:)=bit_moded(:,:); ifft_position(conj_position,:)=conj(bit_moded(:,:)); signal_time = ifft(ifft_position,ifft_length); figure('position',[0 400 400 400],'menubar','none'); subplot(3,1,1) plot(signal_time(:,1),'b'); title('原始单个OFDM符号'); xlabel('Time'); ylabel('Amplitude'); % ==================加循环前缀和后缀================== signal_time_C = [signal_time(end-CP_length+1:end,:);signal_time]; signal_time_C = [signal_time_C; signal_time_C(1:CS_length,:)]; subplot(3,1,2); % 单个完整符号为512+128+20=660 plot(signal_time_C(:,1)); xlabel('Time'); ylabel('Amplitude'); title('加CP和CS的单个OFDM符号'); % =======================加窗======================== signal_window = zeros(size(signal_time_C)); % 通过矩阵点乘 signal_window = signal_time_C.*repmat(rcoswindow(alpha,size(signal_time_C,1)),1,symbol_count); subplot(3,1,3) plot(signal_window(:,1)) title('加窗后的单个OFDM符号') xlabel('Time'); ylabel('Amplitude'); % ===================发送信号，多径信道==================== signal_Tx = reshape(signal_window,1,[]); % 变成时域一个完整信号，待传输 signal_origin = reshape(signal_time_C,1,[]); % 未加窗完整信号 mult_path_am = [1 0.2 0.1]; % 多径幅度 mutt_path_time = [0 20 50]; % 多径时延 windowed_Tx = zeros(size(signal_Tx)); path2 = 0.2*[zeros(1,20) signal_Tx(1:end-20) ]; path3 = 0.1*[zeros(1,50) signal_Tx(1:end-50) ]; signal_Tx_mult = signal_Tx + path2 + path3; % 多径信号 figure('menubar','none') subplot(2,1,1) plot(signal_Tx_mult) title('多径下OFDM信号') xlabel('Time/samples') ylabel('Amplitude') subplot(2,1,2) plot(signal_Tx) title('单径下OFDM信号') xlabel('Time/samples') ylabel('Amplitude') % =====================发送信号频谱======================== % ====================未加窗信号频谱======================= % 每个符号求频谱再平均，功率取对数 figure % 归一化 orgin_aver_power = 20*log10(mean(abs(fft(signal_time_C')))); subplot(2,1,1) plot((1:length(orgin_aver_power))/length(orgin_aver_power),orgin_aver_power) hold on plot(0:1/length(orgin_aver_power):1 ,-35,'rd') hold off axis([0 1 -40 max(orgin_aver_power)]) grid on title('未加窗信号频谱') % ====================加窗信号频谱========================= orgin_aver_power = 20*log10(mean(abs(fft(signal_window')))); subplot(2,1,2) plot((1:length(orgin_aver_power))/length(orgin_aver_power),orgin_aver_power) hold on plot(0:1/length(orgin_aver_power):1 ,-35,'rd') hold off axis([0 1 -40 max(orgin_aver_power)]) grid on title('加窗信号频谱') % ========================加AWGN========================== signal_power_sig = var(signal_Tx); % 单径发送信号功率 signal_power_mut = var(signal_Tx_mult); % 多径发送信号功率 SNR_linear = 10^(SNR/10); noise_power_mut = signal_power_mut/SNR_linear; noise_power_sig = signal_power_sig/SNR_linear; noise_sig = randn(size(signal_Tx))*sqrt(noise_power_sig); noise_mut = randn(size(signal_Tx_mult))*sqrt(noise_power_mut); % noise_sig=0; % noise_mut=0; Rx_data_sig = signal_Tx+noise_sig; Rx_data_mut = signal_Tx_mult+noise_mut; % =======================串并转换========================== Rx_data_mut = reshape(Rx_data_mut,ifft_length+CS_length+CP_length,[]); Rx_data_sig = reshape(Rx_data_sig,ifft_length+CS_length+CP_length,[]); % ====================去循环前缀和后缀====================== Rx_data_sig(1:CP_length,:) = []; Rx_data_sig(end-CS_length+1:end,:) = []; Rx_data_mut(1:CP_length,:) = []; Rx_data_mut(end-CS_length+1:end,:) = []; % =========================FFT============================= fft_sig = fft(Rx_data_sig); fft_mut = fft(Rx_data_mut); % =========================降采样=========================== data_sig = fft_sig(carrier_position,:); data_mut = fft_mut(carrier_position,:); figure scatter(real(reshape(data_sig,1,[])),imag(reshape(data_sig,1,[])),'.') grid on; figure scatter(real(reshape(data_mut,1,[])),imag(reshape(data_mut,1,[])),'.') grid on; % =========================逆映射=========================== bit_demod_sig = reshape(qamdemod(data_sig,16,'OutputType','bit'),[],1); bit_demod_mut = reshape(qamdemod(data_mut,16,'OutputType','bit'),[],1); % =========================误码率=========================== error_bit_sig = sum(bit_demod_sig~=bit_sequence); error_bit_mut = sum(bit_demod_mut~=bit_sequence); error_rate_sig = error_bit_sig/bit_length; error_rate_mut = error_bit_mut/bit_length; rate = [rate; error_rate_sig error_rate_mut] % ========================================================== % ==========================================================