Knowledge-aided parametric adaptive matched filter with automatic combining for covariance estimation

Pu Wang, Hongbin Li, Zhe Wang, Braham Himed

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

3 Scopus citations

Abstract

In this paper, a knowledge-aided parametric adaptive matched filter (KA-PAMF) is proposed that utilizing both observations (including the test and training signals) and a priori knowledge of the spatial co-variance matrix. Unlike existing KA-PAMF methods, the proposed KA-PAMF is able to automatically adjust the combining weight of a priori covariance matrix, thus gaining enhanced robustness against uncertainty in the prior knowledge. Meanwhile, the proposed KA-PAMF is significantly more efficient than its KA non-parametric counterparts when the amount of training signals is limited. One distinct feature of the proposed KA-PAMF is the inclusion of both the test and training signals for automatic determination of the combining weights for the prior spatial covariance matrix and observations. Numerical results are presented to demonstrate the effectiveness of the proposed KA-PAMF, especially in the limited training scenarios.

Original languageEnglish
Title of host publication2014 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2014
Pages6067-6071
Number of pages5
DOIs
StatePublished - 2014
Event2014 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2014 - Florence, Italy
Duration: 4 May 20149 May 2014

Publication series

NameICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings
ISSN (Print)1520-6149

Conference

Conference2014 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2014
Country/TerritoryItaly
CityFlorence
Period4/05/149/05/14

Keywords

  • STAP
  • parametric adaptive matched filter

Fingerprint

Dive into the research topics of 'Knowledge-aided parametric adaptive matched filter with automatic combining for covariance estimation'. Together they form a unique fingerprint.

Cite this