关于盲信号自适应分离中非线性函数的讨论

This paper proposes a new algorithm of blind source separation (BSS). The algorithm can overcome the difficulty known as "the sensors are less than the source signals" and works effectively when the sensors are less. Then, the paper discusses the nonlinear functions used in the new algorithm. A uniform nonlinear function is proposed and some criterion are given to choose its parameters. Finally, some simulations are presented to show the effectness of the algorithm and the correctness of the criterion.