J4 ›› 2012, Vol. 39 ›› Issue (4): 52-56.doi: 10.3969/j.issn.1001-2400.2012.04.010

• 研究论文 • 上一篇    下一篇

利用数据最小冗余拟合的信源数估计方法

杨志伟1;贺顺1,2;张娟1;徐青1
  

  1. (1. 西安电子科技大学 雷达信号处理国家重点实验室,陕西 西安  710071;
    2. 西安科技大学 通信学院,陕西 西安  710054)
  • 收稿日期:2011-05-05 出版日期:2012-08-20 发布日期:2012-10-08
  • 通讯作者: 杨志伟
  • 作者简介:杨志伟(1980-),男,副教授,博士,E-mail: zwyang@mail.xidian.edu.cn.
  • 基金资助:

    国家自然科学基金资助项目(60901066);预研基金资助项目(9140C0101081001);中央高校基本科研业务费和博士点基金资助项目(20090203120006)

Estimation of the number of sources via minimal redundancy fitting

YANG Zhiwei1;HE Shun1,2;ZHANG Juan1;XU Qing1   

  1. (1. National Key Lab. of Radar Signal Processing, Xidian Univ., Xi'an  710071, China;
    2. Commun. and Info. Eng. Collage, Xi'an Univ. of Sci. and Tech., Xi'an  710054, China)
  • Received:2011-05-05 Online:2012-08-20 Published:2012-10-08
  • Contact: YANG Zhiwei

摘要:

针对低信噪比和小样本条件下难以正确估计信源数目问题,提出利用数据最小冗余拟合检测的信源数估计新方法.该方法首先采用子空间跟踪算法估计训练样本集的一维噪声子空间,然后根据对噪声子空间的谱峰搜索结果构造出信号的冗余阵列流形,最后通过检测该流形对阵列数据的最小冗余表示来估计信源数.仿真试验检验了所提方法的有效性.

关键词: 阵列信号处理, AIC准则, MDL准则, 信源数估计

Abstract:

To alleviate the performance deterioration in the estimation of the number of sources with a low signal-to-noise ratio and/or secondary data deficient scenario, a new method based on the minimum redundancy fitting test is presented. The approach is performed in three stages. First, we employ a subspace tracking algorithm to calculate the one-dimensional noise subspace, and then a redundant array steer matrix is obtained based on the spatial spectrum which corresponds to the noise subspace. Redundant steer vectors are removed via minimum redundancy fitting and the number of sources is obtained. The effectiveness of the presented algorithm is verified by numerical simulation.

Key words: array signal processing, AIC criteria, MDL criteria, source number estimation