A known signal detection scheme using rank statistics in multiplicative noise

Multiplicative noise is known to be useful in modeling multipath propagation which is important in the analysis of mobile communication systems. In this paper, nonparametric detection of known signals in multiplicative noise is considered. The locally optimum detector is derived based on signs and ranks of observations for good weak-signal detection performance under any specified noise probability density function. This detector has similarities to the locally optimum detector for known signals in multiplicative noise. The asymptotic performance of this nonparametric detector is as good as that of the locally optimum detector. With approximate score functions, we can construct the locally suboptimum rank detector.