基于PPSO-GRADS算法的盲信号分离

Blind Signal Separation Based on PPSO-GRADS Algorithm

该文将"联姻"策略应用在粒子群算法中,提出一种并行分阶段的基于粒子群优化算法的盲信号分离方法(PPSO-GRADS)。该算法具有收敛速度快,分离精度高的特点。通过仿真证明该算法比未使用"联姻"策略的粒子群算法有更好的性能,在收敛速度和分离效果上比传统的梯度算法,遗传算法都有较明显的改善。

In this paper, allied strategy is introduced into PSO and a parallel and two-stage blind signal separation based on particle swarm optimization （parallel particle swarm optimization-gradient algorithm） is proposed. This algorithm is characteristic of rapid convergent speed and high separation precision. The simulation results show that this algorithm proposed is superior to particle swarm optimization in which allied strategy is not applied, and this algorithm is superior to traditional gradient algorithm and genetic algorithm in convergent speed and separation effect.