基于动态阈值函数的改进小波包遥测信号去噪方法

An Improved Wavelet Packet De-noising of Telemetry Signal Based on Dynamic Threshold Function Method

摘要:

通过遥测信号可以获取飞行器的工作状态和环境数据,为评定性能和故障分析提供依据。本文利用小波包变换能对高频信号分解的优点,针对传统小波包去噪方法的阈值函数在处理既含突变又含平滑信号的情形下,不能在保持良好信号边缘特性与消除恒定偏差的同时,又能避免出现信号原本没有的振荡的问题,提出一种基于动态阈值函数的改进小波包遥测信号去噪方法,通过量化当前处理信号的平滑程度和构造介于硬阈值函数和软阈值函数之间并可切换的动态阈值函数,跟随信号变化进行去噪处理。通过算例仿真分析,验证了本文方法在处理既含突变又含平滑信号时,比传统小波包法具有更好的去噪效果。

Abstract:

The working state and environmental data of the aircraft can be obtained by telemetry signal, which provides a basis for performance evaluation and fault analysis. In this paper, takes the advantage of wavelet packet transforming can decompose the high frequency signal, in terms of the problem when the threshold function of the traditional wavelet packet de-noising method is used to deal with both mutation and smooth signal, it is impossible to maintain good signal edge characteristics and eliminating constant deviation while avoiding the oscillation that the signal does not have, An improved wavelet packet de-noising of telemetry signal based on dynamic threshold function method is proposed. By quantized the smoothness of the current signal and the wavelet packet dynamic threshold function which is between hard threshold function and soft threshold function and can be switched, follow the signal changes to carry out de-noising. Simulation results show that the proposed method has better de-noising effect than the traditional wavelet packet method for the both mutation and smooth signal.