A nonlinear sea clutter analysis using chaotic system

Abstract

Sea clutter refers to the data gathered from backscatters from the sea surface which when illuminated by radar pulses. The amplitude of sea clutter data is very complex. The random like behavior of Sea Clutter amplitude is due to interaction of a large number of degree of freedom from the physical system. The factors affecting the sea clutter are mainly two parameters, Environmental parameters and Radar Parameters. Modeling of sea clutter counting these entire factors is very difficult. It is of great importance to develop a model to characterize sea clutter accurately. So for this reason stochastic process is used to model the complex and random like behavior of sea clutter. A lot of work has been done to fit the amplitude of sea clutter data to various distributions such as K distribution, Rayleigh distribution etc. but this method is not good for detecting the target within the sea clutter. A new approach known as "Chaotic Method" had been developed for modeling and analyzing sea clutter data. A deterministic chaotic system could be used to model the complex, random like behavior of sea clutter using small number of degree of freedom. The advantage of this method is that it may leads to better understanding of process which produces clutter and also to detect the target within the sea clutter. Here we analyze the data which was obtained from Mac Master University and confirmed that the randomness of sea clutter data is generated by deterministic chaos not by stochastic process. The characteristics of chaos that is used to prove the randomness of sea clutter are Lypunov Exponent, Kolmogorov Entropy and Correlation dimension. This paper also verifies the Takens Embedding Theorem which states that dynamical reconstruction of system can be achieved by using single dimension subspace. Although the reconstruction of sea clutter cannot produce a model exactly the same as the original one, it has the same behavior as the original system.