Detection and localization in colored noise via generalized least squares

Mati Wax, Jacob Sheinvald, Anthony J. Weiss

Research output: Contribution to journalArticlepeer-review

Abstract

A method for detection and localization of multiple signals in spatially colored noise by an arbitrary passive sensor array is presented. The method also enables exploitation of prior knowledge that the signals are uncorrelated so as to improve the performance and to allow detection and localization even if the number of signals exceeds the number of sensors. The estimation, based on the generalized least squares criterion, is both consistent and asymptotically efficient. The detection is performed via the minimum description length (MDL) principle and is proved to be consistent. Simulation results confirming the theoretical results are included.

Original languageEnglish
Pages (from-to)1734-1743
Number of pages10
JournalIEEE Transactions on Signal Processing
Volume44
Issue number7
DOIs
StatePublished - 1996
Externally publishedYes

Fingerprint

Dive into the research topics of 'Detection and localization in colored noise via generalized least squares'. Together they form a unique fingerprint.

Cite this