Robust CFAR Detection in Heterogeneous Weibull Background via Bayesian Area Interference Control

Constant false alarm rate (CFAR) detection in heterogeneous Weibull background has always been an important issue in the field of radar target detection. However, interference effects and non-Gaussian characteristics of the background usually lead to inaccurate prediction and complex computation of background level for existing methods, which results in severe degradation of CFAR detection performance. To address the problem of CFAR detection in the heterogeneous Weibull background, two robust CFAR detectors capable of detecting multiple targets accurately and efficiently are developed. Specifically, a Bayesian predictive inference detection model is derived to establish a standard CFAR detection framework. Then, combining the interference control approach, a Bayesian interference control CFAR (BIC-CFAR) detector in the heterogeneous Weibull background is proposed. Next, to alleviate the high computational complexity faced by the BIC-CFAR detector, a Bayesian area interference control CFAR (BAIC-CFAR) detector is designed to predict and compensate interferences simultaneously, resulting in the improvement of anti-interference capability while dramatically reducing computational complexity. Simulations and real-data experiments are conducted in heterogeneous Weibull background, which demonstrates 10% improvement in detection probability and 30% reduction in computational time.