Intelligent CFAR Detector Based on Support Vector Machine

Abstract

In this paper, we propose an intelligent constant false alarm rate detector, which uses support vector machine (SVM) techniques to improve the radar detection performance in different background environments. The proposed detector uses the variability index statistic as a feature to train a SVM and recognizes the current operational environment based on the classification results. The proposed detector has the intelligence to select the proper detector threshold adaptive to the current operational environment. This detector provides a low loss performance in homogeneous backgrounds and also performs robustly in nonhomogeneous environments including multiple targets and clutter edges.

Keywords

• Constant false alarm rate
• support vector machine
• variability index
• automatic censored cell averaging
• greatest of