OFDM ISAC 雷达的一些经典和新型传感算法的比较

%% A comparison of OFDM Radar signal processing methods % Description: This project compares some classical sensing algorithm. % Algorithms in this project: % % 1. 2DFFT % 2. Cyclic Cross Correlation % 3. Super Resolution Estimation: MUSIC and TLS-ESPRIT % % Author: Yunbo HU (SIMIT, UCAS) % GitHub: https://github.com/edenhu1111 clc; clear; addpath('fig'); global c0 fc lambda M N delta_f Ts CPsize %% ISAC Transmitter % System parameters c0 = 3e+8; % velocity of light fc = 30e+9; % carrier frequency lambda = c0 / fc; % wavelength M = 1024; % number of subcarriers N = 15; % number of symbols per subframe delta_f = 120e+3; % subcarrier spacing T = 1 / delta_f; % symbol duration Tcp = T / 4; % cyclic prefix duration Ts = T + Tcp; % total symbol duration CPsize = M / 4; % cyclic prefix length bitsPerSymbol = 2; % bits per symbol qam = 2^(bitsPerSymbol); % 16-QAM modulation % Transmit data data = randi([0 qam - 1], M, N); TxData = qammod(data, qam, 'gray'); % OFDM modulator TxSignal = ifft(TxData, M); % IFFT TxSignal_cp = [TxSignal(M - CPsize + 1: M, :); TxSignal]; % add CP TxSignal_cp = reshape(TxSignal_cp, [], 1); % time-domain transmit signal %% Channel % Sensing Data Generation SNR = 30; r = [30]; % range.(it can be a row vector) v = [20]; %velocity RxSignal = sensingSignalGen(TxSignal_cp, r,v,SNR); k = length(r); %% OFDM Radar Receivers % 1. 2DFFT based (classical method, reference is omitted here) Rx = RxSignal(1:size(TxSignal_cp,1),:); Rx = reshape(Rx,[],N); Rx = Rx(CPsize + 1 : M + CPsize,:); % remove CP and reshape it into a matrix Rx_dem = fft(Rx,M); CIM_2dfft = Rx_dem .* conj(TxData); %elementwise-multiply(equals to match filtering) RDM_2dfft = fft(ifft(CIM_2dfft,M).',10*N); % plot the range doppler map figure(1); range_2dfft = linspace(0,c0/(2*delta_f),M+1); range_2dfft = range_2dfft(1:M); velocity_2dfft = linspace(0,lambda/2/Ts,10*N+1); velocity_2dfft = velocity_2dfft(1:10*N); [X,Y] = meshgrid(range_2dfft,velocity_2dfft); RDM_2dfft_norm = 10*log10( abs(RDM_2dfft)/max(abs(RDM_2dfft),[],'all')); % Maximized Likelihood Estimation(In theory) mesh(X,Y,(RDM_2dfft_norm)); title('2D-FFT based method'); xlabel('range(m)'); ylabel('velocity(m/s)'); savefig('fig/figure1.fig'); % 2. CCC-based (Method proposed by Kai Wu et al.) figure(2); % setting parameters for CCC-based sensing method % please refer to the paper for the meaning of these paramaters mildM = 512; Qbar = 64; mildQ = 128; % CCC [r_cc,RDM] = cccSensing(RxSignal,TxSignal_cp,mildM,Qbar,mildQ); % ccc sensing %plot the range doppler map Tsa = 1/delta_f/M; mildN = floor((length(TxSignal_cp)-Qbar-mildQ)/(mildM - Qbar)); range_ccc = linspace(0,c0/2*Tsa*mildM, mildM+1); doppler_ccc = linspace(0,lambda/(mildM-Qbar)/Tsa/2,10*mildN+1); range_ccc = range_ccc(1:mildM); doppler_ccc = doppler_ccc(1:10*mildN); RDM_norm = 10*log10(abs(RDM)/max(abs(RDM),[],'all')); [X,Y] = meshgrid(range_ccc,doppler_ccc); mesh(X,Y,(RDM_norm)); % plot the range-doppler map title('CCC based method'); xlabel('range(m)'); ylabel('velocity(m/s)'); savefig('fig/figure2.fig'); % 3. Super resolution sensing method % 3.1 MUSIC based (a time consuming but precise method) CIM = Rx_dem .*conj(TxData); [P_music_range,P_music_velo] = MUSICforOFDMsensing(CIM,k); % plot the MUSIC power spectrum figure(3); title('MUSIC for OFDM sensing'); subplot(1,2,1); plot(linspace(0,100,length(P_music_range)),abs(P_music_range)/max(abs(P_music_range))); ylabel('Pmusic'); xlabel('range(m)'); ylim([10^-3,1]); title('MUSIC for range estimation'); subplot(1,2,2); plot(linspace(0,100,M),abs(P_music_velo)/max(abs(P_music_velo))); ylabel('Pmusic'); xlabel('velocity(m/s)'); ylim([10^-3,1]); title('MUSIC for velocity estimation'); savefig('fig/figure3.fig'); % 3.2 ESPRIT based method [range,velocity] = ESPRITforOFDMsensing(CIM,k); fprintf('The estimation result of TLS-ESPRIT is :\n'); fprintf('Range = %f\n',range); fprintf('Velocity = %f\n',velocity);