Nonparametric permutation test versus optimum parametric test for radar detection

In this paper, we analyze a detector based on the optimum permutation test, applied to nonparametric radar detection which provides good performance without a large computational effort. We compare it with the parametric test in the Neyman-Pearson sense. We also show the characteristic of detectability of the optimum permutation test versus parametric one under Gaussian noise environments and different types of target models (nonfluctuating, Swerling I and Swerling II). The detection probability versus signal-to-noise ratio is calculated by Monte-Carlo simulations for different parameter values (pulse number N, reference samples M and false alarm probability P_fa)