alternating_optimization_交替优化的源代码_

%disp('Running alternating optimization algorithm') % r=1; % iteration index % Initialize reflection matrix theta beta_vec = ones(M,1); % Fixed to 1 for now as in the paper theta_vec = 2*pi*rand(M,1); % Uniformly randomized from 0 to 2*pi theta_mat= diag(beta_vec.*exp(1i*theta_vec)); H = Ht'*(theta_mat')*Hr + Hd; % Check rank criterion for feasbility of the initial theta choice while ~(rank(H) == N_users) % if infeasible choice, randomize and check again %disp('infeasible initial choice of theta, .. reselecting ..') theta_vec = 2*pi*rand(M,1); % Uniformly randomized from 0 to 2*pi theta_mat= diag(beta_vec.*exp(1i*theta_vec)); H = Ht'*(theta_mat')*Hr + Hd; end cvx_status = 'nothing'; % initialize while (frac_error > eps_iter) && ~contains(cvx_status,'Infeasible','IgnoreCase',true) % if mod(r,1e2)==0 % %disp(['Iteration r =' num2str(r)]) % end H = Ht'*(theta_mat')*Hr + Hd; % ==== Optimize W while fixing theta ==== BS Transmit Beamforming %disp('Active Beamformer Design') [W, tau, INTERF, cvx_status, cvx_optval] = iter_opt_prob_1(H,sigma_2,SINR_target,int_users_matrix); if cvx_optval==Inf %disp('Infeasible .. passing this iteration') continue end %disp(['CVX Status: ' cvx_status ', CVX_optval = ' num2str(10*log10(cvx_optval*1000)) ' dBm']) %disp(['CVX Status: ' cvx_status ', CVX_optval = ' num2str(10*log10(trace(W'*W)*1000)) ' dBm']) frac_error = abs(obj_last - cvx_optval)/obj_last *100; obj_last = cvx_optval; achieved_SINR = zeros(1,N_users); % Actual achieved SINR for k = all_users achieved_SINR(k) = (norm((H(:,k)')*W(:,k)))^2/(norm(INTERF(:,k)))^2; end % ==== Optimize theta while fixing W ==== IRS Reflection Matrix % (P4') in paper %disp('Passive Beamformer Design') [V, a_aux, a, b, R, desired, interference, SINR_CONSTR, cvx_status, cvx_optval] = iter_opt_prob_2(W, Ht,Hr,Hd,sigma_2,SINR_target,int_users_matrix); %disp(['CVX Status: ' cvx_status]) if ~contains(cvx_status,'Infeasible','IgnoreCase',true) %disp('Running Gaussian Randomization') [U,D] = eig(full(V)); % Eigenvalue Decomposition if rank(full(V)) == 1 v_bar = U*sqrt(D); theta_vec = angle(v_bar(1:M)/v_bar(M+1)); v = exp(-1i*theta_vec); theta_mat = diag(v); else % Apply Gaussian Randomization num_rands = 1e3; % number of randomizations % Generate Gaussian random vector ~ CN(0, I) %gpudev = gpuDevice(); %reset(gpudev); r_vec_matrix = (1/sqrt(2))*((mvnrnd(zeros(M+1,1),eye(M+1),num_rands) + 1i * mvnrnd(zeros(M+1,1),eye(M+1), num_rands)).'); %gpuArray() v_bar_matrix = U*sqrt(D)*r_vec_matrix; best_index = 0; best_value = -1e8; %v_bar_matrix = exp(1i*2*pi*rand(M+1,num_rands)); for randmzn_index = 1:num_rands v_bar_vec = v_bar_matrix(:,randmzn_index); V_rand = v_bar_vec*(v_bar_vec'); [~, ~, constr_value] = sinr_CONSTRAINT(V_rand, b, R, SINR_target, sigma_2, all_users, int_users_matrix); % Check feasibility and best value feasibility_check = prod( constr_value >= 0 ); better_value_check = (sum(constr_value) > best_value); if feasibility_check && better_value_check best_index = randmzn_index; best_value = sum(constr_value); end end if best_index ~= 0 % select best v_bar that maximizes SINR_CONSTR v_bar = v_bar_matrix(:,best_index); theta_vec = angle(v_bar(1:M)/v_bar(M+1)); v = exp(-1i*theta_vec); theta_mat = diag(v); else cvx_status = 'Infeasible'; end %disp(['CVX Status after randomization: ' cvx_status]) end end % % Increment iteration index % r = r+1; end