ML-ZF算法源码程序 - matlab算法设计

下载频道> 资源分类> matlab源码> 算法设计> ML-ZF算法源码程序

标题：ML-ZF算法源码程序
分享到：

所属分类: 算法设计资源类型：文件大小: 11.44 KB 上传时间: 2016-01-27 19:31:29 下载次数: 56 资源积分：1分提供者: 马云 ML-ZF算法源码程序

内容：
ML-ZF算法源码程序，程序员在编程的过程中可以参考学习使用，希望对IT程序员有用，此源码程序简单易懂、方便阅读，有很好的学习价值！

clear all;clc

N = 2; % Number of Transmit antennas

M = 2; % Number of Receive antennas

EbNoVec = 5:3:11; % Eb/No in dB

modOrd = 2; % constellation size = 2^modOrd

numSym = N; % number of symbols

% Seed states for repeatability

rand('twister', 12345); randn('state', 98765);

% Get all bit combinations for ML receiver

bits = de2bi(0:2^(modOrd*N)-1, 'left-msb')';

% Split them per Transmit antenna

b = zeros(N, modOrd, length(bits));

for i = 1:length(bits)

b(:, :, i) = reshape(bits(:,i), modOrd, N)';

end

% Preallocate variables for speed

dist = zeros(length(bits), 1);

[BER_ZF, BER_MMSE, BER_ML] = deal(zeros(1, length(EbNoVec)));

% Create QPSK mod-demod objects

hMod = modem.pskmod('M', 2^modOrd, 'SymbolOrder', 'gray', 'InputType', 'bit');

hDemod = modem.pskdemod(hMod);

% Set up a figure for visualizing BER results

h = gcf; grid on; hold on;

set(gca,'yscale','log','xlim',[EbNoVec(1)-0.01, EbNoVec(end)],'ylim',[1e-4 1]);

xlabel('Eb/No (dB)'); ylabel('BER'); set(h,'NumberTitle','off');

set(h, 'renderer', 'zbuffer'); set(h,'Name','Spatial Multiplexing');

title('2x2 Uncoded QPSK System');

% Loop over selected EbNo points

for idx = 1:length(EbNoVec)

nErrs_zf = 0; nErrs_mmse = 0; nErrs_ml = 0;

nBits = 0;

while ( ((nErrs_zf < 100) || (nErrs_mmse < 100) || (nErrs_ml < 100)) ...

&& (nBits < 1e4))

% Create array of bits to modulate

msg = randint(modOrd, numSym, 2);

% Modulate data

source = modulate(hMod, msg);

% Split source among N transmitters (symbol-wise)

Tx = reshape(source, N, numel(source)/N); clear source;

% Flat Rayleigh Fading - independent links

RayleighMat = (randn(M, N) + sqrt(-1)*randn(M, N))/sqrt(2);

% Calculate SNR from EbNo

snr = EbNoVec(idx) + 10*log10(modOrd);

% Add channel noise power to faded data

r = awgn(RayleighMat*Tx, snr); clear Tx;

r_store = r;

% Assume perfect channel estimation

H = RayleighMat;

% Zero-Forcing SIC receiver

E_zf = zeros(modOrd, numSym); k = zeros(N, 1);

% Initialization

G = pinv(H);

[val, k0] = min(sum(abs(G).^2,2));

% Start Zero-Forcing Nulling Loop

for n = 1:N

% Find best transmitter signal using minimum norm

k(n) = k0;

% Select Weight vector for best transmitter signal

w = G(k(n),:);

% Calculate output for transmitter n and demodulate bitstream

y = w * r;

E_zf(:, k(n):N:end) = reshape(demodulate(hDemod, y), modOrd, numSym/N);

% Subtract effect of the transmitter n from received signal

z = modulate(hMod, demodulate(hDemod, y));

r = r - H(:, k(n))*z;

% Adjust channel estimate matrix for next minimum norm search

H(:, k(n)) = zeros(M, 1);

G = pinv(H);

for aa = 1:n

G(k(aa), :) = inf;

end

[val, k0] = min(sum(abs(G).^2,2));

end

% Restore variables for next receiver

H = RayleighMat; r = r_store;

% MMSE SIC receiver

E_mmse = zeros(modOrd, numSym); k = zeros(N, 1);

% Initialization

G = inv(H'*H + N/(10^(0.1*snr))*eye(N)) * H';

[val, k0] = min(sum(abs(G).^2,2));

% Start MMSE Nulling Loop

for n = 1:N

% Find best transmitter signal using Min Norm

k(n) = k0;

% Select Weight vector for best transmitter signal

w = G(k(n),:);

% Calculate output for transmitter n and demodulate bitstream

y = w * r;

E_mmse(:, k(n):N:end) = reshape(demodulate(hDemod, y), modOrd, numSym/N);

% Subtract effect of the transmitter n from received signal

z = modulate(hMod, demodulate(hDemod, y));

r = r - H(:, k(n))*z;

% Adjust channel estimate matrix for next min Norm search

H(:, k(n)) = zeros(M, 1);

G = inv(H'*H + N/(10^(0.1*snr))*eye(N)) * H';

for aa = 1:n

G(k(aa), :) = inf;

end

[val, k0] = min(sum(abs(G).^2,2));

end

% Restore variables for next receiver

H = RayleighMat; r = r_store;

% ML receiver

for i = 1:2^(modOrd*N)

% Signal constellation for each bit combination

sig = modulate(hMod, b(:, :, i)').';

% Distance metric for each constellation

dist(i) = sum(abs(r - H*sig).^2);

end

% Get the minimum

[notUsed, val] = min(dist);

E_ml = b(:,:,val)'; % detected bits

% Collect errors

nErrs_zf = nErrs_zf + biterr(msg, E_zf);

nErrs_mmse = nErrs_mmse + biterr(msg, E_mmse);

nErrs_ml = nErrs_ml + biterr(msg, E_ml);

nBits = nBits + length(msg(:));

end

% Calculate BER for current point

BER_ZF(idx) = nErrs_zf./nBits;

BER_MMSE(idx) = nErrs_mmse./nBits;

BER_ML(idx) = nErrs_ml./nBits;

% Plot results

semilogy(EbNoVec(1:idx), BER_ZF(1:idx), 'r*', ...

EbNoVec(1:idx), BER_MMSE(1:idx), 'bo', ...

EbNoVec(1:idx), BER_ML(1:idx), 'gs');

legend('ZF-SIC', 'MMSE-SIC', 'ML');

drawnow;

end

% Draw the lines

semilogy(EbNoVec, BER_ZF, 'r-', EbNoVec, BER_MMSE, 'b-', ...

EbNoVec, BER_ML, 'g-');

hold off;

openfig('spatMuxResults.fig');

文件列表（点击上边下载按钮，如果是垃圾文件请在下面评价差评或者投诉）:

ML-ZF算法源码程序/
ML-ZF算法源码程序/MMSE_ZF.m
ML-ZF算法源码程序/spatMuxResults.png

关键词：算法 源码 程序

matlab源码下载排行

回到顶部

友情链接

matlab源码网站源码编程语言网络编程手机通讯编程 KUNAG 官网 KUNAG 电商更多

: 联系方式| 版权声明| 招聘信息| 广告服务| 银行汇款| 法律顾问| 兼职技术| 付款方式| 关于我们|; 网站客服程序员兼职招聘 411241577@qq.com; 沪ICP备19040327号-3; 公安备案号：沪公网安备 31011802003874号; 库纳格流体控制系统（上海）有限公司版权所有; Copyright © 1999-2014, GUSUCODE.COM, All Rights Reserved

关键词： 算法 源码 程序

相关推荐

matlab源码下载排行

友情链接

关键词：算法源码程序