1.算法运行效果图预览

2.算法运行软件版本

matlab2022a

3.部分核心程序

.......................................................................
            %子载波分配
            [~,pow2] = func_subcarriers_capacity(Ptotal, ch, N_subcarrier, K, noise, gamma);

            %功率分配
            tic
            ianp      = func_power(ch,pow2,N_subcarrier,K,Ptotal,noise,gamma); 
            time_end2 = toc;

            Avg_time2(ij1) = Avg_time2(ij1) + time_end2;

            for i=1:K
                pow1_water(i) = func_waterfilling(shenp(i),pow1(i,:).*ch(i,:)/noise)/N_subcarrier;
                pow2_water(i) = func_waterfilling(ianp(i),pow2(i,:).*ch(i,:)/noise)/N_subcarrier;
            end;

            cap2=cap2+sum(pow1_water);
            cap1=cap1+sum(pow2_water);

            if  ij2 == 1 
                cap_m1 = cap_m1 + pow1_water;
                cap_m2 = cap_m2 + pow2_water;
            end
            norm1 = norm1 + norm(pow2_water/sum(pow2_water) - gamma/sum(gamma), inf);
            norm2 = norm2 + norm(pow1_water/sum(pow1_water) - gamma/sum(gamma), inf);
        end

        if  ij2 == 1
            cap_m1 = cap_m1/(N_ch*MTKL);
            cap_m2 = cap_m2/(N_ch*MTKL);
            
            figure(5);
            bar([gamma/sum(gamma); cap_m2/sum(cap_m2); cap_m1/sum(cap_m1)]', 'grouped');
            legend('Gamma方法', 'LINEAR方法', 'ROOT-FINDING方法');
        end;
    end
    cap1_mean(ij1)=cap1/(N_ch*MTKL);
    cap2_mean(ij1)=cap2/(N_ch*MTKL);
    
    norm1_mean(ij1) = norm1/(N_ch*MTKL);
    norm2_mean(ij1) = norm2/(N_ch*MTKL);
end

 
figure(1)
plot(diff_Vuser,cap1_mean,'-bs',...
    'LineWidth',1,...
    'MarkerSize',6,...
    'MarkerEdgeColor','k',...
    'MarkerFaceColor',[0.9,0.0,0.0]);
hold on
plot(diff_Vuser, cap2_mean,'-r>',...
    'LineWidth',1,...
    'MarkerSize',6,...
    'MarkerEdgeColor','k',...
    'MarkerFaceColor',[0.9,0.9,0.0]);
grid on
xlabel('用户数')
ylabel('容量 (bit/s/Hz)')
legend('LINEAR', 'ROOT-FINDING'); 
hold off

Avg_time  = Avg_time/(N_ch*MTKL);
Avg_time2 = Avg_time2/(N_ch*MTKL);

figure(3);
semilogy(diff_Vuser,Avg_time2,'-bs',...
    'LineWidth',1,...
    'MarkerSize',6,...
    'MarkerEdgeColor','k',...
    'MarkerFaceColor',[0.9,0.0,0.0]);
hold on
semilogy(diff_Vuser,Avg_time,'-r>',...
    'LineWidth',1,...
    'MarkerSize',6,...
    'MarkerEdgeColor','k',...
    'MarkerFaceColor',[0.9,0.9,0.0]);
grid on
xlabel('用户数')
ylabel('平均仿真时间 (s)')
legend('LINEAR', 'ROOT-FINDING');
0040

4.算法理论概述

在OFDM通信系统中，资源分配是一项关键任务，它涉及将可用的频谱资源和功率分配给不同的子载波，以实现高效的数据传输。为了降低计算复杂度并提高系统性能，低复杂度的资源分配算法成为研究的焦点之一。OFDM（正交频分复用）是一种广泛用于无线通信的调制技术，它将高速数据流分成多个低速子流，并将它们调制在不同的正交子载波上。这样可以减少多径干扰，提高频谱利用率。

考虑一个具有N个子载波的OFDM系统。信道的频率响应可以表示为H(k)，其中k表示子载波索引。发送的数据符号为X(k)，接收端接收到的信号为Y(k)。接收信号可以表示为：