一种优化动量因子的非线性主重量分析算法摘 要 盲源分离算法收敛速度和稳态误差存在冲突,针对这一问题提出了通过引入动量项并且对动量因子进行优化的方法来使收敛速度和稳态误差之间冲突达到最小
该算法通过优化的动量项使非线性主成分分析的代价函数最速下降,从而使算法收敛速度加快
仿真实验证明了本文算法在平稳与非平稳环境下具有比 LMS 和 RLS 更好的分离性能
关键词 动量项 , 非线性主成分分析 , 自适应 , 盲源分离A new NPCA algorithm for blind source separationAbstractThis paper addresses the problem of blind source separation (BSS) and presents an optimum momentum factor which makes the nonlinear principal component analysis (NPCA) cost function descend in the fastest way
By using the momentum item in self-stabilized NPCA algorithm, the new algorithm minimums the contradiction between convergence speed and steady-state error
Simulations show that the new algorithm has faster convergence than the existingleast-mean-square(LMS)algorithms and recursive least-squares (RLS) algorithm for BSS in