Signal detection in additive-multiplicative non-Gaussian noise using higher order statistics

This paper addresses the problem of signal detection in additive-multiplicative non-Gaussian noise, when the moment and cumulant functions are used to describe the random process. The new approach for statistical hypothesis testing is proposed. This approach is based on the development of the polynomial decision rules and new moment quality criterion statistical hypothesis testing. It is shown that nonlinear signals processing and taking into account the cumulants of third and higher order is allowed us to describe non-Gaussian process and increase the affectivity of signal processing.