Radar detection of signals with unknown parameters in K-distributed clutter

The detection of signals with unknown parameters in correlated K-distributed noise, using the generalised Neyman-Pearson strategy is considered. The a priori uncertainty on the signal is removed by performing a maximum likelihood estimate of the unknown parameters. The resulting receivers can be regarded as a generalisation of the conventional detector, but for a zero-memory nonlinearity depending on the amplitude probability density function of the noise as well as on the number of integrated pulses. It is shown that the performance for uncorrelated observations is unaffected by the specific signal pattern, but depends only on the signal-to-noise ratio; moreover, the effect of the clutter correlation on the performance can be accounted for simply by a detection gain. A performance assessment, carried out by computer simulation, shows that the proposed receivers significantly outperform conventional ones as the noise amplitude probability density function markedly deviates from the Rayleigh law. It also shows that the generalised Neyman-Pearson strategy is a suitable means of circumventing the uncertainty on wanted target echos since the operating characteristics of the receivers for the case of signals with unknown parameters closely follow those of the receiver for a completely known signal.