摘要
设计的自适应神经网络噪声抵消系统不需要关于输入信号的先验知识,非线性映射能力强,具有自学习能力、计算量小、实时性好。利用该系统对含噪声的非线性信号建模,达到消除噪声的目的。通过LMS算法,对不同信噪比(SNR)的含噪信号进行滤波。仿真结果表明,该滤波器能有效地抑制噪声。
Based on the character about adaptive Neural Network filtering needn't previous information of input single, have better ability of nonlinear mapping and self-study,a novel adaptive noise cancellation system is designed.The system has the nature of a little quantity in its calculation and good for processing in real_time,can using in nonlinear modeling of noise single,and gets to noise cancellation.So a method are designed using LMS algorithm,and filtering about single of different SNR.It is shown that the filter is good for restraining noise by simulation.
出处
《计算机工程与应用》
CSCD
北大核心
2005年第28期65-67,共3页
Computer Engineering and Applications
基金
湖南省教育厅科学研究项目(编号:03C499)
关键词
噪声
神经网络
噪声抵消
自适应滤波
noise,Neural Network,noise cancellation,adaptive filtering