摘要
针对含噪环境下的盲源分离问题,将一种稳健的含噪条件下的白化预处理方法应用于FastICA算法中,提出了一种改进的FastICA算法。实验仿真结果表明:该算法的抗噪声性能比经典的FastICA算法和RobustICA算法有了较大地改善,而运算量基本不变。
An improved FastICA algorithm is proposed aiming at the problem of blind source separa- tion in noisy circumstances, which combines the robust whitening algorithm in noisy circumstances and FastICA algorithm. Simulation results show that compared with the classic FastICA algorithm and the RoubustICA algorithm, the proposed algorithm has greatly-improved anti-noise performance without increasing computation quantity.
出处
《信息工程大学学报》
2013年第6期708-712,共5页
Journal of Information Engineering University
作者简介
吴微（1981-）,女,博士生,主要研究方向为信号分析和盲信号处理,E-mail：wuwei_0930@163.com.
彭华（1973-）,男,教授,博士生导师,主要研究方向为通信信号处理和软件无线电.
