Radar detection prediction in sea clutter using the compound K-distribution model

In the paper the radar detection of targets in sea clutter modelled by the compound K-distribution is examined from a statistical detection viewpoint. Optimum performance is determined for both fixed threshold and ideal constant false alarm rate (CFAR) receivers for targets modelled by Swerling cases 0, 1 and 2. In particular, the performance of dual threshold binary integrators is assessed. The compound K-distribution sea clutter model is not only a good fit to observed amplitude statistics over a wide range of conditions, but also allows the temporal and spatial correlation characteristics to be accurately modelled. The modelling of pulse-to-pulse correlation characteristics is very important for accurate performance prediction when pulse-to-pulse integration is used. The use of the compound K-distribution model not only gives more accurate results than models based on amplitude distribution alone, but also gives a new insight to the detection characteristics of different radar detection radar detection systems over a wide range of sea conditions.