摘要
对基于最大信噪比的盲源分离算法进行了修正,解决了原算法在一些情况下失效的问题,并且比较分析了以上算法和全局最优盲源分离算法的分离性能。仿真结果表明,修正了的基于最大信噪比的盲源分离算法和全局最优盲源分离算法经常在分离性能上很相近,且都解决了修正前算法在一些情况下失效的问题。
This paper revises the blind source separation algorithm based on maximum signal noise ratio and solves the problem of losing the effect under some conditions of the original algorithm. This paper also compares and analyzes the separation performances of the above two and the algorithm based on global optimal property. Simulations show that the separation performances of the revised algorithm based on maximum signal noise ratio and the algorithm based on global optimal property are always similar, and they both solves the problem of losing the effect under some conditions of the original algorithm.
出处
《电脑与信息技术》
2009年第1期19-21,共3页
Computer and Information Technology
关键词
最大信噪比
滑动平均
特征向量
全局最优
maximum signal noise ratio
moving average
eigenvector
global optimal property
