计算机工程与应用 ›› 2012, Vol. 48 ›› Issue (5): 146-148.

• 数据库、信号与信息处理 • 上一篇    下一篇

一种低信噪比下的特征向量类信源数估计算法

李 媛1,王 畅2,郭建新1,刘尚峰1,3   

  1. 1.空军工程大学 电讯工程学院,西安 710077
    2.华北水利水电学院,郑州 450011
    3.空军雷达学院,武汉 430019
  • 收稿日期:1900-01-01 修回日期:1900-01-01 出版日期:2012-02-11 发布日期:2012-02-11

Novel source enumeration algorithm at low SNR based on eigenvectors

LI Yuan1, WANG Chang2, GUO Jianxin1, LIU Shangfeng1,3   

  1. 1.Institute of Telecommunication Engineering, Air Force Engineering University, Xi’an 710077, China
    2.North China University of Water Resources and Electric Power, Zhengzhou 450011, China
    3.Air Force Radar Academy, Wuhan 430019, China
  • Received:1900-01-01 Revised:1900-01-01 Online:2012-02-11 Published:2012-02-11

摘要: 经典的特征值类信源数估计算法在低信噪比、少快拍数条件下的估计性能急剧下降,针对该问题,提出了一种新的信源数估计算法。该算法利用采样协方差矩阵的特征向量对信噪比不敏感的特性来构造判决变量,根据改进的预测描述长度(PDL)准则来实现对信源数的有效估计,理论分析和仿真实验证明了所提算法的有效性。

关键词: 信源数估计, 特征向量, 低信噪比

Abstract: Performance of classical source estimation algorithm drops quickly at low SNR and small snapshot conditions. To deal with this problem, a novel source enumeration algorithm is proposed. The approach employs sample covariance matrix’s eigenvectors which are not sensitive to SNR as decision value. Improved Predictive Description Length(PDL) criterion is adopted to enumerate source number effectively. Theoretical analysis and simulation results demonstrate the validity of the method.

Key words: source enumeration, eigenvector, low signal-to-noise ratio