Enhancing Distributed Radar Detection Performance With Reliable Communications

Abstract

Despite their potential to significantly enhance target detection capability, distributed radar networks, with or without a central coordinator, have not been sufficiently examined. In this paper, we propose a generalized likelihood ratio test (GLRT)-based constant false alarm rate (CFAR) detector for distributed detection in a radar network, wherein only neighboring nodes require sharing their GLRT values. We obtain the analytical expression for the local GLRT and employ distributed average consensus-based estimation to reach the global metric in the case of noise-free communications links. The exact analytical expressions for the local probability of false alarm (PFA) and probability of detection (PD) are derived and used to obtain the exact analytical expression for the global PFA as well as an approximate expression for the global PD. In the case of noisy communications links, we demonstrate the effects of quantization and different modulation techniques, such as chirp and phase shift keying, on the global detection probability. Our results show that the proposed detector performs significantly better than the non-CFAR energy detector (ED) and provides performance comparable to that of the non-CFAR matched filter (MF) detector. It is shown that the performance of the chirp and binary phase shift keying (BPSK) modulations is similar but better than that of quadrature phase shift keying (QPSK). Our results also show that the uniform quantization of GLRT values with only 4 bits is sufficient to enable each node to achieve performance similar to the global performance achieved in the case of error-free communications links.