Performance Analysis of non-Gaussian Sea Clutter Modelling Using Compound Inverted Exponentiated Rayleigh Distribution

Abstract

Radar echo modelling is a major step in the development of robust detection schemes. Compound Gaussian models are largely used to model non-Gaussian radar clutter and provided good results, compound models are characterized by two components speckle and texture, where the speckle component follows Rayleigh law. In this dissertation, the modelling performance of the compound inverted exponentiated Rayleigh (CIER) distribution is tested. A comparative study conducted using the real data of high-resolution sea clutter collected using IPIX radar by McMaster University, Canada. The fitted probability density function (PDFs) and the complimentary cumulative density functions (CCDFs) of the CIER, and the well-known compound K, the compound inverse Gaussian (CIG) distributions are compared with the real PDF and CCDF. The results obtained validate the ability of the CIER distribution to model high-resolution sea clutter.