Statistical characteristics of large-scale bottom reverberation

Abstract

Simulation results from a statistical characterization of large-scale bottom reverberation from broadband signals are presented. The seafloor profile is parametrized via the surface root-mean-square height and correlation length. A connection between these two parameters and the statistical moments of the backscattered signal is established. It is shown that the variance of reverberation increases proportionally to the seafloor root-mean-square height. On the contrary, the kurtosis (or coefficient of excess) of reverberation appears to be independent of changes in roughness height and related to the terrain's smoothness. In particular, it is shown that kurtosis variations are proportional to changes in the seafloor correlation length.