Comparison between different HOS-based tests for detection of ship-radiated signals in non-Gaussian noise

Abstract

Conventional signal processing algorithms and detection criteria, optimised in presence of Gaussian noise, may degrade their performances in non-Gaussian environments. Higher Order Statistics (HOS) theory is a powerful means both for characterizing non-Gaussian noise and, then, for designing efficient and robust signal detectors. In particular, a method for detecting signals in additive independent non-Gaussian background noise have been investigated, analysed and compared with the bispectrum-based Hinich test and with a conventional spectrum-based detection criterion. In order to compare their respective performances, the different approaches have been applied on real underwater acoustic data, recording the passage of a target ship, in presence of background shipping traffic noise. >