一种改进的自适应噪声抵消系统算法研究  

Algorithm Study on an Improved Adaptive Noise Cancellation System

作　　者：伍彩云[1] 翁晶晶 WU Caiyun;WENG Jingjing(Shenyang Ligong University,Shenyang 110159,China)

机构地区：[1]沈阳理工大学装备工程学院,沈阳110159

出　　处：《沈阳理工大学学报》2022年第3期1-7,共7页Journal of Shenyang Ligong University

基　　金：国家自然科学基金资助项目(61603264)。

摘　　要：自适应噪声抵消系统能够从含噪环境中检测和提取到有用信号,对噪声信号进行抵消。本文对自适应噪声抵消系统中最小均方(LMS)算法进行研究,针对其存在收敛速度和稳态误差间不协调的缺点,提出一种新的基于激活函数Softsign的变步长LMS算法,设计了自适应噪声抵消系统仿真平台,利用该平台验证所提出算法的去噪效果。结果表明,该算法在收敛速度和抗噪性能方面都优于传统LMS算法和已有变步长LMS算法。该仿真平台具有添加信号、叠加噪声、去噪处理、波形显示以及计算信噪比等功能,新算法的提出与仿真平台的设计为自适应噪声抵消技术的研究提供了新的解决方案。Adaptive noise cancellation system can detect and extract the useful signal from the noisy environment and cancel the noisy signal.In this paper, the least mean square(LMS)algorithm of adaptive noise cancellation system is studied, and the convergence speed and steady-state error of LMS algorithm are incompatible, a new variable-step LMS algorithm based on Softsign function is proposed.Adaptive noise cancellation system simulation platform is designed and used to verify the denoising effect of the proposed algorithm.The results show that the algorithm outperforms both the traditional LMS algorithm and the existing variable-step LMS algorithm in terms of convergence speed and noise-resistant performance.The simulation platform has the functions of adding signal, superimposing noise, denoising, displaying waveform and calculating signal-to-noise ratio, etc.The proposed new algorithm and the design of the simulation platform provide a new solution for the research of adaptive noise cancellation technology.

关 键 词：自适应噪声抵消系统 变步长最小均方算法 收敛速度 抗噪性能 

分 类 号：TN911.72[电子电信—通信与信息系统]