电子学报 ›› 2015, Vol. 43 ›› Issue (3): 483-488.DOI: 10.3969/j.issn.0372-2112.2015.03.010

• 学术论文 • 上一篇    下一篇

冲激噪声环境下基于最大相关熵准则的韧性子空间跟踪新算法

张金凤1,2, 邱天爽1, 李森3   

  1. 1. 大连理工大学电子信息与电气工程学部, 辽宁大连 116024;
    2. 深圳大学深圳市现代通信与信息处理重点实验室, 广东深圳 518060;
    3. 大连海事大学信息科学技术学院, 辽宁大连 116026
  • 收稿日期:2013-10-30 修回日期:2014-03-30 出版日期:2015-03-25
    • 作者简介:
    • 张金凤 女,河北定州人,1979年9月出生,深圳大学讲师,大连理工大学在读博士生.主要研究方向为非高斯信号分析与处理. E-mail:zhangjf@szu.edu.cn;邱天爽 男,江苏海门人,1954年8月生,大连理工大学教授,博士生导师.从事信号信息处理方面的教学科研工作.在国内外学术期刊与会议上发表论文200余篇,曾获教育部科学技术二等奖等多项科技奖励. E-mail:qiutsh@dlut.edu.cn
    • 基金资助:
    • 国家自然科学基金 (No.61172108,No.61139001,No.61301228); 中央高校基本科研业务费资助项目 (No.3132014212)

A Robust PAST Algorithm Based on Maximum Correntropy Criterion for Impulsive Noise Environments

ZHANG Jin-feng1,2, QIU Tian-shuang1, LI Sen3   

  1. 1. Faculty of Electronic Information and Electrical Engineering, Dalian University of Technology, Dalian, Liaoning 116024, China;
    2. Shenzhen Key Lab of Advanced Communications and Information Processing, Shenzhen University, Shenzhen, Guangdong 518060, China;
    3. Information Science Technology College, Dalian Maritime University, Dalian, Liaoning 116026, China
  • Received:2013-10-30 Revised:2014-03-30 Online:2015-03-25 Published:2015-03-25
    • Supported by:
    • National Natural Science Foundation of China (No.61172108, No.61139001, No.61301228); Fundamental Research Funds for the Central Universities (No.3132014212)

摘要:

本文采用最大相关熵准则(MCC)对投影近似子空间跟踪(PAST)算法中基于最小平方误差(MSE)准则的目标函数进行修正,推导出适用于冲激噪声环境的韧性投影近似子空间跟踪新算法(MCC_PAST算法).对两种冲激噪声模型包括,稳定分布噪声模型和混合高斯噪声模型环境下的时变波达方向估计问题的实验仿真表明,与基于韧性M估计的PAST改进算法(RLM_PAST算法)相比,MCC_PAST算法可以自适应地调整核长,对于阵列接收数据的变化体现出更好的适应性.

关键词: 投影近似子空间跟踪(PAST), 相关熵, 冲激噪声, M估计

Abstract:

By employing the maximum correntropy criterion(MCC)based cost function in projection approximation subspace tracking(PAST)algorithm, we deduce the MCC_PAST algorithm which can be utilized for the projection approximation subspace tracking under impulsive noise environments.We examine the performance of our proposed algorithm in the time-varying DOA(Direction Of Arrival)estimation problem under two types of impulsive noise environments, including the alpha-stable noise environments and the contaminated Gaussian noise environments.By adjusting the kernel parameter adaptively in the algorithm, MCC_PAST exhibits superiority over the robust M-estimate based PAST algorithm, especially when there are sudden changes in the array outputs.

Key words: projection approximation subspace tracking(PAST), correntropy, impulsive noise, M-estimate

中图分类号: