Implementing the Method for Adaptive Nonlinear Filtration of Non-Gaussian Clutter in a Coherent Receiver Band Path and Estimating Its Efficiency

Abstract

The necessity of investigating the possibilities of amplitude adaptive nonlinear filtration of clutter is stemming from the existence of a large number of electromagnetic radiation sources of both natural and artificial origins. Most often, such clutter is of a non-Gaussian type and can significantly reduce the detection efficiency of received signals in radar and telecommunication systems. Theoretically, one of effective ways to deal with non- Gaussian clutter can be the known method of asymptotically optimal amplitude nonlinear suppression. However, this method is not suitable for being widely used in practice, as it has a number of mandatory conditions that cannot normally be met in reality. To obtain the optimum amplitude characteristic of a nonlinear converter, it is necessary to have a priori information about the clutter amplitude probability density, which cannot be obtained in most cases. Also, the observation conditions and the clutter environment change with time; therefore, the statistical characteristics of the received clutter cannot be determined in the majority of real cases. In addition, in view of the limitation imposed on a small signal-to-clutter ratio, in the absence of clutter expected by the detection system, the useful signal will be suppressed, due to which the detection efficiency will tend to zero. Thus, such method is not an adaptive one.

At present, a search of methods for solving these problems is a topical issue, which is the subject of this article. The possibility of implementing the proposed method of amplitude adaptive nonlinear clutter suppression in the receiver’s coherent bandpass path and the efficiency of this method are investigated. The efficiencies of the asymptotically optimal nonlinear conversion and of the proposed adaptive clutter suppression method are compared on a model of radar reflections from a wavy sea surface. It is shown that the proposed adaptive clutter suppression method is almost equal in efficiency to the asymptotically optimal nonlinear converter, but unlike the latter, it does not need a priori information about the clutter distribution and does not impose limitations on the small signal-to-clutter ratio.