Performance analysis of TM-CFAR detection in time diversity systems for pulse-to-pulse correlated targets in presence of clutter edges and outlying targets: A comparison through extensive simulations

In automatic radar detection, performance of time diversity systems is proved to be sensitive to how the multiple-pulse sample echoes, from commonly encountered nohomogeneities, are accumulated within the same receiver. Based upon the ldquonon conventional time diversity systemrdquo(NCTDS), we derive exact expressions for the probabilities of false alarm and detection of a pulse-to pulse partially correlated chi-square target with 2K degrees of freedom for the trimmed mean constant false alarm rate (TM-CFAR) detector. Because of the complex mathematics induced by the ldquoconventional time diversity systemrdquo (CTDS), detection performance of this system is carried out using Monte Carlo simulations. The extensive simulation results show that, while the NCTDS achieves a better nonuniform clutter resilience than the CTDS, the latter outperforms the former in multiple target situations. Hence, when both nonhomogeneities are concomitantly present, the two systems performances become nearly similar.