LFM信号的分数阶傅里叶域自适应滤波算法研究  

Adaptive Filtering Algorithm for LFM Signal in fraction Fourier domain

作　　者：安澄全[1] 郝冬艳 

机构地区：[1]哈尔滨工程大学,通信与信息工程学院,黑龙江哈尔滨150001

出　　处：《信号处理》2013年第1期54-59,共6页Journal of Signal Processing

基　　金：国家自然科学基金资助项目(61275082)

摘　　要：对于线性调频信号(LFM)的滤波,采用处理平稳信号的方法对其滤波往往得不到很好的效果。本文利用了线性调频信号在分数阶傅里叶变换域上具有很好的时频聚焦性的特点,来实现信号在分数阶傅里叶域的自适应滤波,自适应滤波算法采用改进的步长LMS方法,对传统的LMS算法做出了改进,算法中步长处理中引入了一个限制因子,可以较好地解决算法收敛速度和稳态失调量之间的矛盾。仿真结果表明,此算法在处理分数阶域的LFM信号滤波比传统的LMS算法有较好的滤波效果。It is not very good results to achieve the filter of the LFM （ linear frequency modulation signal） signal by the method of deal with stationary signal. In this paper, an adaptive method for filtering the LFM signal is proposed, using the characteristic of good energy concentration on LFM signal in fraction Fourier transform domain. Adaptive filter algorithm a- dopted improve step-size LMS algorithm, it made improvements to the traditional LMS algorithm, the algorithm added to a limiting factor, in order to resolve conflicts between Convergence speed and steady-state mis-adjustment, and to improve ill- tering performance in fraction Fourier transform domain. LMS algorithm was used well, because its Robustness was very good and its structure was simple, it was seen to be easy in computation and simple in implementation, Simulation results represent that the proposed LMS algorithm has good performance than the traditional LMS algorithm in fraction Fourier transform domain.

关 键 词：线性调频信号 分数阶傅里叶变换域 自适应滤波算法 改进步长 限制因子 

分 类 号：TN911.71[电子电信—通信与信息系统]