Non-Parametric Detection of Targets with Dynamic Fluctuation Behaviour

This is the authors' abstract. We don't add key points for this paper.

This study presents an approach to radar detection that addresses the possible challenges posed by future targets utilizing cloaking technologies and electronic countermeasures to decorrelate the scattered signal. The work focuses on the adaptive selection of appropriate Coherent Processing Intervals (CPI), formulating it as a multi-hypotheses test problem. A compound detector is derived that combines the output of several normalized fixed-CPI detectors using a point-wise maximum operation. This, with the goal of enhancing target detection probability amidst unknown fluctuations. A conservative upper bound is given for the probability of false alarm, and the performance of the detector is evaluated through Monte Carlo analysis. The results highlight the ability of the compound detector to consistently outperform fixed-CPI detectors for fluctuating targets while mitigating potential performance losses for non-fluctuating targets. Overall, the proposed conceptually simple detector strikes a balance between robustness and detection performance, suggesting its utility for detecting targets with dynamic fluctuation behaviour.