资源简介
基于块稀疏信号的重构算法。稀疏贝叶斯学习算法。
代码片段和文件信息
function res = SBI(paras)
% res = SBI(paras)
%
% SBI(paras) performs DOA estimation using Sparse Bayesian Inference
%
% Input:
% paras.Y: M * T matrix sensor measurements at all snapshots
% paras.A: M * N matrix columns are the steering vectors for different directions
% paras.B: M * N matrix columns are derivatives of the steering vectors wrt. different directions
% paras.sigma2: initialization of noise variance
% paras.alpha: initialization of alpha
% paras.beta: initialization of beta
% paras.rho: rho
% paras.resolution: grid resolution for the directions
% paras.maxiter: maximum iteration
% paras.tol: stopping criterion
% paras.isKnownNoiseVar: true if known variance false if unknown
% paras.K: number of sources
% Output:
% res.mu: mean estimation
% res.Sigma: variance estimation
% res.sigma2: estimated noise variance
% res.sigma2seq: estimated noise variance at all iterations
% res.alpha: reconstructed alpha
% res.beta: reconstructed beta
% res.iter: iteration used in the algorithm
% res.ML: maximum likelihood function value at all iterations
%
% Written by Zai Yang 19 Jul 2011
% reference: Zai Yang Lihua Xie and Cishen Zhang
% “Off-grid direction of arrival estimation using sparse Bayesian inference“
eps = 1e-16;
Y = paras.Y;
A = paras.A;
B = paras.B;
[M T] = size(Y);
N = size(A 2);
alpha0 = 1 / paras.sigma2;
rho = paras.rho / T;
beta = paras.beta;
alpha = paras.alpha;
r = paras.resolution;
maxiter = paras.maxiter;
tol = paras.tolerance;
if isfield(paras ‘isKnownNoiseVar‘) && ~isempty(paras.isKnownNoiseVar)
isKnownNoiseVar = paras.isKnownNoiseVar;
else
isKnownNoiseVar = false;
end
if isKnownNoiseVar
a = 1;
b = T * M * paras.knownsigma2;
else
a = 1e-4;
b = 1e-4;
end
if isfield(paras ‘K‘) && ~isempty(paras.K)
K = paras.K;
else
K = min(T M-1);
end
idx = [];
BHB = B‘ * B;
converged = false;
iter_beta = 1;
iter = 0;
ML = zeros(maxiter1);
alpha0seq = zeros(maxiter1);
while ~converged
iter = iter + 1;
Phi = A;
Phi(:idx) = A(:idx) + B(:idx) * diag(beta(idx));
alpha_last = alpha;
C = 1 / alpha0 * eye(M) + Phi * diag(alpha) * Phi‘;
% Sigma = diag(alpha) - diag(alpha) * Phi‘ / C * Phi * di
- 上一篇:鸡群优化算法和鸟群算法源代码
- 下一篇:基于半色调技术的图像影藏
相关资源
- Pattern Recognition and Machine Learning(高清
- MATLAB 编程 第二版 Stephen J. Chapman 著
- 均值滤波和FFT频谱分析Matlab代码
- 《MATLAB扩展编程》代码
- HDB3码、AMI码的MATLAB实现
- 3点GPS定位MATLAB仿真
- MATLAB数字信号处理85个实用案例精讲入
- matlab从入门到精通pdf94795
- 欧拉放大论文及matlab代码
- 跳一跳辅助_matlab版本
- 全面详解LTE MATLAB建模、仿真与实现
- MIMO-OFDM无线通信技术及MATLAB实现_孙锴
- MATLAB Programming for Engineers 4th - Chapman
- matlab 各种谱分析对比
- 分数阶chen混沌matlab程序
- 基于粒子群算法的非合作博弈的matl
- MATLAB车流仿真 包括跟驰、延误
- matlab空间桁架计算程序
- 基于MATLAB的图像特征点匹配和筛选
- DMA-TVP-FAVAR
- GPS信号的码捕获matlab代码.7z
- 一维光子晶体MATLAB仿真代码吸收率折
- newmark法源程序
- 传统关联成像、计算鬼成像matlab
- pri传统分选算法
- 摆动滚子推杆盘形凸轮设计
- 医学图像重建作业matlab源码
- Matlab实现混沌系统的控制
- 检测疲劳驾驶
- Matlab锁相环仿真-Phase Locked Loop.rar
评论
共有 条评论