宽带信号噪声抑制技术研究

发布时间：2018-04-16 00:41

本文选题：宽带信号 + 降噪　；参考：《电子科技大学》2014年硕士论文

【摘要】：宽带信号无论在生活中还是军事上的应用都越来越广,但是在应用过程中,不可避免的会在信号中混入噪声,严重的情况下甚至会让信号面目全非。所以宽带信号降噪就成为了一个很有研究价值和意义的课题。本文在对目前主流降噪算法进行了全面了解的基础上,选择了常应用于窄带信号进行降噪的小波阈值降噪算法,对宽带线性调频(LFM)信号进行降噪的尝试。并对这种算法从两个方面进行了改进。进行了仿真对比,分析出每一种方法的优缺点。最后将两者结合成一种降噪方法,对宽带LFM信号的降噪取得了较好的效果。本文的主要工作有:1.介绍小波分析的相关基本理论,由连续小波变换,二进小波变换,小波级数三个层次,递进的说明小波分析对信号进行分解的原理,并详细阐述了多分辨分析和mallat快速算法的过程。2.介绍了小波阈值降噪算法的原理和流程,对小波基的选取、阈值函数的选取和阈值估计,以及降噪以后效果的评估体系进行了详细说明,并针对宽带LFM信号选取了合适的参数。3.通过对常见的小波阈值降噪算法进行仿真对比,分析出这种方法的局限性,在低频噪声过滤和时间分割两个方面进行了算法的改进,通过实验仿真证明这种改进有较好的效果。
[Abstract]:Wideband signal is more and more widely used in life and military affairs, but in the process of application, it is inevitable to mix noise in the signal, and even make the signal completely different in serious cases.So wideband signal noise reduction has become a research value and significance of the subject.On the basis of a comprehensive understanding of the current mainstream denoising algorithms, the wavelet threshold de-noising algorithm, which is often applied to narrow band signals, is selected in this paper to try to reduce the noise of wideband linear frequency modulation (LFM) signals.The algorithm is improved from two aspects.The advantages and disadvantages of each method are analyzed.Finally, the two methods are combined into a noise reduction method, and good results are obtained for wideband LFM signal de-noising.The main work of this paper is: 1.This paper introduces the basic theory of wavelet analysis, which consists of three levels: continuous wavelet transform, dyadic wavelet transform and wavelet series.The process of multi-resolution analysis and mallat fast algorithm. 2. 2.The principle and flow of wavelet threshold denoising algorithm are introduced. The selection of wavelet basis, the selection of threshold function and threshold estimation, as well as the evaluation system of effect after denoising are described in detail, and the appropriate parameter. 3 is selected for wideband LFM signal.The limitation of this method is analyzed by comparing the common wavelet threshold denoising algorithms. The algorithm is improved in the aspects of low frequency noise filtering and time segmentation. The experimental results show that the improved algorithm has good results.
【学位授予单位】：电子科技大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：TN911.4

【参考文献】