Coherent radar detection of targets against a combination of K-distributed and Gaussian clutter

This paper introduces the mathematical model of radar echoes generated by the mixture of coherent K-distributed and Gaussian distributed clutter sources. The optimum strategy to detect an a priori known target against the above mentioned disturbance and thermal noise is also presented, based on the likelihood ratio test. A Monte Carlo simulation has been performed to evaluate the receiver operating characteristics (ROCs). The probability of detection for a given false alarm probability has been calculated as a function of signal to noise ratio, K-distributed clutter to noise ratio, Gaussian distributed clutter to noise ratio, number of integrated pulses, order parameter /spl upsi/, correlation lengths of K-clutter and Gaussian disturbance. Moreover the performance of such a detector is compared with that of the detectors designed to take into account only one type of disturbance, either the K-distributed clutter or the Gaussian clutter.