基于小波变换的微弱信息检测研究

2018-10-12 05:48:38王月
现代电子技术 2018年19期

王月

摘 要: 传统微弱信息检测方法不能过滤纯噪信号,常因纯噪信号幅值、强度过大,造成信息检测结果失真。为解决此问题,设计基于小波变换的微弱信息检测模型。通过微弱信息小波分解尺度确定、小波作用阈值改进、微弱信息小波变换系数重构,完成微弱信息的小波变换去噪;通过小波Duffing振子重构信号检测、微弱信息待测信号频率确定、微弱信息幅值测量,完成基于小波变换微弱信息检测模型的搭建。模拟该模型的使用环境,设计对比实验。结果表明,应用基于小波变换微弱信息检测模型后,纯噪信号幅值与强度明显降低。

关键词: 小波变换; 微弱信息检测; 去噪分析; Duffing振子; 信号检测; 幅值测量

中图分类号: TN911?34 文献标识码: A 文章编号: 1004?373X(2018)19?0071?04

Abstract: The traditional weak information detection method can′t filter the pure noise signal, and may cause the distortion of the information detection result due to the high amplitude and strong intensity of the pure noise signal. Therefore, a weak information detection model based on wavelet transform is designed to solve the above problem. The wavelet transform denoising of the weak information is accomplished by means of determination of wavelet decomposition scale of weak signal, improvement of the wavelet action threshold and reconstruction of the wavelet transform coefficients of weak information. The weak information detection model based on wavelet transform is built by means of detection of wavelet Duffing oscillator reconstruction signal, determination of signal frequency of weak information, and amplitude measurement of weak information. The service environment of the model is simulated. The contrast experimental results show that the amplitude and intensity of the pure noise signal are significantly reduced by using the weak information detection model based on wavelet transform.

Keywords: wavelet transform; weak information detection; denoising analysis; Duffing oscillator; signal detection; amplitude measurement

0 引 言

傳统微弱信号检测方法利用电子学、物理学和信息学基础理论,以小电容、弱磁、微振动等信号作为检测对象,利用相应传感技术,将检测信号转变为恒定的电压、电流或电量。通过检测恒定电力常量的相关参数,进而检测出信号的微弱量。有时待检测信号过于微弱,常被强度过大的纯噪信号掩盖,造成检测结果与真实结果间差距过大[1?2]。为从纯噪信号中提取出完整的微弱信息,引入小波变换原理。小波变换降低了信号检测过程中所设置的目标门限,使检测所得的微弱信号精度得到了有效提升。微弱信号检测是一项在噪声中发现有用信号的新兴技术,在研究过程中通过待测信号与噪声幅度、相位、变化频率等特征的不同,确定噪声的产生原因及规律,并根据确定结果,研究去除纯噪信号的有效方法[3?4]。……

登录APP查看全文