摘要
小波分析是近年来发展起来的一种有效信号处理技术,但由于其时频分解尺度是按二进制变化的,所以在高频段其频率分辨率差,而在低频段其时间分辨率较差。小波包分析是更加精细的信号处理方法,它将频带进行多层次划分,对低频和高频部分同时进行分解,不同于小波分析之处在于把细节部分也回归地作为镜像滤波器的输入,产生一系列小波包的集合,然后应用某种代价函数来选择最佳子集。研究结果表明采用小波包方法比小波分析法能够更加有效地去除轮速信号中的各种干扰。
The wavelet analysis is an efficient signal processing technology which developed in late years, but because its time-frequency decomposing scale varies according to binary system, its resolving power of frequency is poor at its high frequency moment, its resolving power of time is poor at its lower frequency moment.The wavelet packets analysis is a more elabrate method of signal processing, it partitions the frequency band on multilayer, and decomposes lower frequency and high frequency at the same time, different from the wavelet analysis, The wavelet packets analysis makes the detail into the input of mirror filter by regression, which produces the collection of a series of wavelet packets, then the best subset is seleted through applying some cost function.the researching results show that the wavelet packets analysis can more effectively eliminate all noises of wheel signal compare to the wavelet analysis.
出处
《微计算机信息》
2010年第8期22-24,共3页
Control & Automation
基金
安徽省高校自然科学基金项目(KJ2009B248Z)
关键词
轮速信号
小波分析
小波包分析
wheel speed signal
the wavelet analysis
the wavelet packets analysis
