一种采用云自适应粒子群算法的盲源分离

One Kind of Blind Source Separation Applying for Cloud Adaptive Particle Swarm Optimization Algorithm

针对大多盲源分离算法全局收敛性能不理想,收敛速度慢的缺陷,借鉴自适应粒子群算法的思想,利用云模型中云滴的随机性和稳定倾向性特点,提出一种云理论的自适应粒子群(CAPSO)盲源分离算法,以分离信号的峭度为目标函数,用自适应调整策略把粒子群分为三个子群,根据云方法修改普通子群的惯性权重,使惯性权重随着适应度值自适应调整。仿真结果表明,改进算法能完成含噪信号分离,并且有效地避免了早熟收敛,较基本PSO提高了全局搜索能力和收敛速度,分离效果好。

To overcome the shortcomings of the many blind source separation algorithm methods,such as the global convergence performance,and have many problems such as Slow convergence speed.One kind of blind source separation applying for Cloud adaptive Particle swarm optimization algorithm,was proposed.CAPSO is based on both the idea of PSO and the properties of randomness and stable tendency of a normal cloud model.it takes the kurtosis of mixtures as a contrast function.Based on the Adaptive adjustment strategy,the particle swarm is divided into three populations.It modifies the inertia weight using a cloud method,the inertia weight depends on the fitness adjustment adaptively.Experimental results show CAPSO can separate signal completing the gauss noise,can efficiently alleviate the problem of premature convergence and improve the global convergence ability and enhance the rate of convergence than PSO.