Two dirac-deltas detector analysis for target detection in non-gaussian sea clutter

Tomasz Gorski (1, 2) , Adam Kawalec (2) , Witold Czarnecki (2) , Jean-Marc Le Caillec (1) , Laurent Lecornu (1)

1 ITI - Département Image et Traitement Information
2 MUT - Institute of Radioelectronics (Military University of Technology)

Space-Time Adaptive Processing (STAP) is a well known technique used for dealing with clutter in order to detect moving targets. This technique was derived under assumption, that clutter has Gaussian characteristics. Unfortunately when dealing with sea clutter, Gaussian assumption is no longer valid [1]. This causes increased number of false alarms. In this paper we present improved detector to deal with non-Gaussian clutter. Detector was derived from Generalized Likelihood Ratio Test (GLRT), assuming Spherically Invariant Random Process (SIRP) as a model for the clutter. Resulting detector was named Two Dirac-Deltas (TDD) detector and it has additional parameter (Δ) in comparison to classical STAP. Based on simulations, it is shown that it is crucial to choose Δ parameter appropriately.

Mots clés: Space-Time Adaptive Processing (STAP); Gaussian; Generalized Likelihood Ratio Test (GLRT)

Domaines: Traitement du signal et de l'image [eess.SP]

Tomasz Gorski, Adam Kawalec, Witold Czarnecki, Jean-Marc Le Caillec, Laurent Lecornu. Two dirac-deltas detector analysis for target detection in non-gaussian sea clutter. SPS 2007 : Signal Processing Symposium 24-26 may, Jachranka, Pologne, May 2007, Jachranka, Poland. ⟨hal-02122091⟩