Abstract: In order to solve blind source separation without any information of the number of source signals,a new algorithm based on hierarchical genetic was proposed. Control layer and parameter layer represent the available number of rows and the concrete value of the available rows of separation matrix respectively. Non-Gaussian was used to measure the independence of the stochastic signals. The kurtosis was maximized/minimized by searching a linear combination of observation signals, and then the independent variables were separated. Computer simulation verifies the validity of the proposed algorithm.

摘要： 为了解决在没有源信号个数信息条件下的盲信号分离问题,设计了一种基于递阶遗传算法的盲信号分离算法。用控制层、参数层分别表示分离阵有效行数和有效行数的具体数值,将非高斯性作为随机信号相互独立的度量,通过寻求观测信号的一个线性组合,使其峭度最大化或最小化来分离独立分量。计算机仿真验证了所提算法的有效性,对未知源个数盲信号处理有一定的指导意义。

关键词: 未知源信号个数, 递阶遗传算法, 峭度, 盲信号分离, unknown number of sources, hierarchical genetic algorithm, kurtosis, blind source separation