参数自适应混沌粒子群算法在盲源分离中的应用

Application of Adaptive Parameter Chaotic Particle Swarm Optimization Algorithm for Blind Source Separation

独立分量分析(ICA)是盲源信号分离中应用最为广泛技术,其应用过程需要对目标函数进行优化,传统粒子算法(PSO)对其进行优化时,存在易陷入局部最优、稳定性差等缺陷,针对此问题,提出采用参数自适应混沌粒子群算法对ICA进行优化.首先采用对PSO的参数进行自适应调整,提高粒子的搜索能力,然后对粒子群进行混沌扰动,提高算法收敛速度.仿真结果表明,使用参数自适应混沌粒子群算法可以有效解决ICA的目标函数优化问题,极大提高了盲源信号的分离效果.

Independent component analysis （ICA） is a blind source separation technology, and in its application process the objective function needs to be optimized, the traditional algorithm of particle （PSO） easily falls into local optimization, instability and other defects. In order to solve this problem, ICA is optimized by the parameter adaptive chaos particle swarm optimization algorithm. Firstly, PSO parameter are adaptive adjusted to improve the search ability of particle, and the particle is chaos disturbed to improve the convergence rate. The results show that the proposed method has solved the ICA objective function optimization problems, and greatly improve the blind source separation effect.