J4 ›› 2012, Vol. 39 ›› Issue (1): 28-33.doi: 10.3969/j.issn.1001-2400.2012.01.006

• Original Articles • Previous Articles     Next Articles

Research on Bayesian radar adaptive detection

ZHOU Yu;ZHANG Linrang;LIU Nan   

  1. (National Key Lab. of Radar Signal Processing, Xidian Univ., Xi'an  710071, China)
  • Received:2010-11-15 Online:2012-02-20 Published:2012-04-06
  • Contact: ZHOU Yu E-mail:zhouyu@mail.xidian.edu.cn

Abstract:

We consider the problem of detecting a signal of interest in the presence of Gaussian noise with the unknown covariance matrix (CM). The traditional approach relies on modeling CM as a deterministic parameter, and its maximum likelihood (ML) estimation is derived when designing the adaptive detector. The ignorance of prior distribution incurs performance loss when there are only a few training data. In this paper, a different approach is proposed which models CM as a random parameter with inverse Wishart distribution. Under this assumption, the maximum a-posteriori (MAP) estimation of CM is derived. The MAP estimate is in turn used to yield the Bayesian version of the Rao and Wald detector. And the importance of the a priori knowledge can be tuned through the scalar variable. The devised detectors remarkably outperform the non-Bayesian Rao and Wald test in the presence of strongly heterogeneous scenarios (where a very small number of training data are available).

Key words: Bayesian approach, Rao test, Wald test, heterogeneous environment

CLC Number: 

  • TN957