ON THE SIMULATION OF CORRELATED K-DISTRIBUTED RANDOM CLUTTER

This paper discusses methods which can be employed in the simulation of correlated K-distributions which are of importance in the prediction of radar clutter. The results of an exact method, based on the integration of a Langevin equation, are compared with an approximate method in which uncorrelated gamma-distributed cross-section contributions are passed through a linear filter selected to give the correct autocorrelation properties. The exact method obviously satisfies all the criteria for K-distributed correlated noise, however it suffers from the disadvantage, at present, of being restricted to one-dimensional fluctuations having a single exponential autocorrelation function. On the other hand, the approximate method has the advantage of being applicable to arbitrary clutter spectra in either one or two dimensions at the expense of only representing the second-order correlation moments precisely.