基于改进粒子群的盲源分离算法研究

A novel blind source separation method based on improved particle swarm optimization

简要地介绍了盲源分离的基本理论,针对独立分量分析传统的优化算法易于陷入局部最优、收敛精度低的缺点,提出了一种基于改进型粒子群的盲源分离算法,将独立分量分析算法与改进的粒子群算法相结合,以负熵作为目标函数.采用这种改进的粒子群算法对分离矩阵进行调整使各个信号分量之间独立,完成对瞬时混合信号的盲分离.实验信号的分离仿真结果表明,该算法能够有效地完成混叠信号的分离.同时,在与传统的盲源分离算法进行对比中,体现出了更高的分离精度和稳定的性能.

The basic theory of blind source separation is introduced briefly. Traditional optimization algorithm carried out by independent component analysis method is easy to fall into partial optimum value, and the convergence precision is low. In view of these disadvantages, a blind source separation method based on an improved algorithm is put forward. It combines the independent component analysis algorithm and the improved particle swarm optimization algorithm, adopts negentropy as the target function, and optimizes the separation matrix by the improved particle swarm optimization algorithm so as to make each signal component independent, and therefore accomplishes blind source separation of the instantaneous mixed signals. The simulation result indicates that the improved algorithm can effectively separate the mixed signals. And compared with the traditional blind source separation algorithm, the improved algorithm represents higher separation precision and stable performance.