Constant False Alarm Detection Algorithm Based on KL Scattering

With the continuous development of the radar field, millimeter-wave radar target detection, as a major tool, faces important challenges in improving detection performance. Especially in some application scenarios, due to multitarget interference and other reasons, the detection performance of the traditional variability index (VI) constant false alarm rate (CFAR) (VI-CFAR) algorithm decreases significantly in the case where both side windows contain interfering targets. To solve this problem, this paper introduces an innovative algorithm, Kullback–Leibler Divergence and Otsu’s Method Enhanced VI-CFAR (KLOVI-CFAR), to better adapt to the multitarget background environment. By combining the KL scattering and Otsu method, we realize the adaptive rejection of outliers in the reference window and further automatically select the detection algorithm adapted to the processed background environment. The results of simulation experiments verify the excellent detection performance of KLOVI-CFAR in multitarget environments with interfering targets in both side windows. The algorithm not only effectively improves the detection capability and antijamming but also shows good detection performance in homogeneous environment and clutter edge cases. The research results in this paper provide a useful reference for improving the radar target detection algorithm.