瞬时频率计算方法的比较研究和改进

发布时间：2018-01-05 05:08

本文关键词：瞬时频率计算方法的比较研究和改进　出处：《四川大学学报(自然科学版)》2014年06期 　论文类型：期刊论文

【摘要】：希尔伯特-黄变换是目前最广泛使用的提取信号瞬时频率时频分析方法,但在求取瞬时频率过程存在端点效应问题.针对这个问题,本文从解析信号的瞬时频率定义着手,介绍了Bedrosian等式和Nuttall定理的限制条件,分析了希尔伯特变换端点效应的产生原因,研究了基于经验调幅调频分解求取瞬时频率的方法及原理.在此基础上比较了希尔伯特变换、Teager能量算子和直接正交法在求取瞬时频率的优缺点.通过实例验证可以得出以下结论:(1)希尔伯特变换解调信号端点效应明显;(2)Teager能量算子法易受信号形式和噪声影响,误差较大;(3)直接正交法易受信号极值点影响,在极值点处会产生瞬时频率突变.针对直接正交法存在的问题,提出了一种改进算法,仿真结果表明:改进算法的效果较为理想,误差较小,能够解决直接正交法求取瞬时频率在极值点处发生突变的问题.
[Abstract]:Hilbert-Huang transform is the most widely used time-frequency analysis method for extracting the instantaneous frequency of signals, but there exists the problem of endpoint effect in the process of obtaining the instantaneous frequency. Starting with the definition of instantaneous frequency of analytical signals, this paper introduces the limiting conditions of Bedrosian equation and Nuttall theorem, and analyzes the causes of the endpoint effect of Hilbert transform. The method and principle of calculating instantaneous frequency based on empirical AM decomposition are studied, and the Hilbert transform is compared. The advantages and disadvantages of Teager energy operator and direct orthogonal method in calculating instantaneous frequency can be obtained. The following conclusions can be drawn by example: 1) the endpoint effect of Hilbert transform demodulation signal is obvious; The energy operator method is easy to be affected by the signal form and noise, and the error is large. 3) the direct orthogonal method is easy to be affected by the signal extremum point, and the instantaneous frequency change will occur at the extremum point. In view of the problems existing in the direct orthogonal method, an improved algorithm is proposed. The simulation results show that the improved algorithm has better effect and less error, and it can solve the problem that the instantaneous frequency changes at the extremum by the direct orthogonal method.
【作者单位】：海军航空工程学院飞行器检测与应用研究所;
【基金】：总装备部武器装备预研基金(9140A27020212JB14311)
【分类号】：TN911.6
【正文快照】： 1引言非平稳信号的瞬时频率是刻画信号振荡模式的一种重要属性,如何有效的提取信号的瞬时频率是近年来研究的热点问题.目前最为广泛使用的是希尔伯特-黄变换(Hilbert-Huang Trans-form,HHT)时频分析方法,其核心包括两个过程[1]:经验模式分解(Empirical Mode Decomposi-tion,EM

【相似文献】