卢鹏1,
张杨勇2,
张刚1
1.重庆邮电大学通信与信息工程学院??重庆??400065
2.武汉船舶通信研究所??武汉??430079
基金项目:国家自然科学基金(61701067, 61771085, 61671095)，重庆市教育委员会科研基金(KJ1600427, KJ1600429)
详细信息
作者简介:罗忠涛：男，1984年生，讲师，硕士生导师，研究方向为统计信号处理与数字图像处理
卢鹏：男，1994年生，硕士生，研究方向为低频噪声分析与低频通信信号处理
张杨勇：男，1983生年，高级工程师，研究方向为低频通信技术与信号处理
张刚：男，1976生年，副教授，硕士生导师，研究方向为微弱信号检测与混沌信号处理
通讯作者:罗忠涛 luozt@cqupt.edu.cn
中图分类号:TN911计量
文章访问数:1008
HTML全文浏览量:407
PDF下载量:39
被引次数:0
出版历程
收稿日期:2018-06-22
修回日期:2018-12-14
网络出版日期:2018-12-24
刊出日期:2019-05-01
Adaptive Design of Limiters for Impulsive Noise Suppression
Zhongtao LUO1,,,Peng LU1,
Yangyong ZHANG2,
Gang ZHANG1
1. School of Communication and Information Engineering, Chongqing University of Posts and Telecommunications, Chongqing 400065, China
2. Wuhan Maritime Communication Research Institute, Wuhan 430079, China
Funds:The National Natural Science Foundation of China (61701067, 61771085, 61671095), The Scientific Research Foundation of the Chongqing Education Committee (KJ1600427, KJ1600429)
摘要
摘要:针对脉冲型噪声的抑制问题，该文提出一种自适应的限幅器设计方法。该方法以效能函数为指标，采用自适应搜索算法，自动寻找削波器和置零器的最佳门限，且能适用于未知噪声分布的情形。首先分析了效能与非线性函数的关系，给出关键的优化问题。然后考虑到效能函数计算复杂，提出基于线搜索的自适应设计算法。其次针对未知分布情况，考虑非参数化的概率密度估计，该算法能够稳健运行且基本取得最优设计效果。最后，结合两种非高斯噪声和实测大气噪声数据仿真，结果表明：该文方法可自适应寻找最佳门限，使削波器和置零器效能达到最佳；当噪声分布未知时，该文方法无需假设噪声模型，可与非参数化概率密度估计方法结合，取得最优检测效果。
关键词:非线性处理/
效能函数/
自适应优化/
削波器/
置零器
Abstract:An adaptive method of limiter design is proposed to suppress impulsive noise. With a purpose of maximizing the efficacy function, the proposed method searches for optimal thresholds of clipper and blanker, via adaptive line search. Firstly, based on analysis on the relationship between the efficacy and the nonlinearity, the key problem of optimization is proposed. Then, since the calculation of efficacy is hard, an adaptive algorithm based on linear search approach is developed based on linear search to optimize the efficacy. Considering the noise distribution is unknown, the proposed method employs the nonparametric kernel density estimation and works robustly in the presence of estimation error. Finally, numeric simulations demonstrate that the proposed method can obtain the optimal performance of clippers and blankers successfully. In the processing of real atmospheric noise from unknown distribution, the proposed method achieves the best detection performance when combining nonparametric kernel density estimation approach.
Key words:Nonlinear processing/
Efficiency function/
Adaptive optimization/
Clipper/
Blanker
PDF全文下载地址:
https://jeit.ac.cn/article/exportPdf?id=14acd2ca-11ec-4c8d-9634-53f0e315daa4
