摘要
盲信号处理方法中常忽略噪声的影响 ,而实际问题中噪声的影响是存在的 .本文主要讨论了在协方差矩阵未知的加性高斯噪声中混合系数的盲估计问题 .本文以最大似然估计为基础 ,提出一种求解参数的最优化算法 ,给出了混合矩阵和协方差矩阵的计算式 .采用高斯混合模型 (GMM)来逼近源信号的概率密度函数 ,简化了算法中的积分 ,导出了一种基于EM算法的迭代式 .仿真表明 ,算法不仅能稳定收敛 。
Generally, some methods of blind signal processing ignore noise, however, noise affects the performance of algorithms, especially seriously in some areas. This paper provides solutions to the problem that mixing matrix is estimated blindly in Gaussian noise with unknown covariance. Based on Maximum Likelihood estimation, the equations are given for solving the mixing matrix and covariance matrix. Gaussian Mixture Model (GMM) is used to approximate the pdf of sources and results in a practical EM algorithm. Computer simulation shows that this algorithm is convergent and has good performance in low SNR.
出处
《电子学报》
EI
CAS
CSCD
北大核心
2003年第7期974-976,共3页
Acta Electronica Sinica
关键词
高斯噪声
盲信号处理
EM算法
高斯混合模型
Algorithms
Approximation theory
Computer simulation
Gaussian noise (electronic)
Mathematical models
Matrix algebra
Maximum likelihood estimation
Probability density function
Signal processing
