Signal Detection in Clutter Using Maximum Entropy Pdf Estimation Based on Fractional Moments

In this paper, we introduce a new method to improve the detection performance of weak unknown radar signals in the presence of unknown clutter. We use maximum entropy (MAXENT) probability density function (PDF) estimation with a new approach based on a few sampled fractional moments (FM). These moments; i.e, their fractional orders, are obtained from the observed sample variates. Using the fractional moments instead of the integer moments the estimated PDF is quite close to the true PDF. The test statistics is a fractional polynomial of very low order of the received samples