Radar Clutter Research And Simulation

The radar clutter emulate is an indispensable part in the radar signal stimulation. In this paper, the clutter models and simulation algorithms at present are summarized. Based on typical models, clutters are generated on computer so as to study the feasibility of simulation algorithms.The mechanism of clutter is complex, so the clutter is described as a random process. Among all the characters of clutter, amplitude distribution andspectrum (or correlation functions) are extremely important. Rayleigh log-normal Weibull and k-type distribution are usually used to describe the amplitude distribution of cluster models. And there are some power spectrum models such as Gaussian exponential and the combination of both.To simulate the clutter, first we should search for a accurate and fast algorithm which could generate series of sequence with a jointly specified amplitude distribution and correlation function. There are two typical kinds of methods for clutter simulation: one is called Zero Memory Nonlinearity (ZMNL) transformation; the other is based on the theory of Spherically Invariant Random Processes (SIRP). Each of them has advantages and disadvantages. We choose one according to the clutter model and our needs.The detailed procedures of the two methods applied to the specified models are given, and the ways to solve the problems which we may meet with are provided. Modern techniques of spectral estimation are used for modeling and simulation, it is one of our specialities. The results should be tested to check their availability and reliability whenever they are generated by models or simulation. Some common ways of parameter extraction and test are summarized in the paper. And the way of testing the power spectrum is also concluded.