Combatting inherent vulnerabilities of CFAR algorithms and a new robust CFAR design

Abstract

A current trend in radar technology is automatic detection and tracking systems. An integral part of these automatic systems is the CFAR (constant False Alarm Rate) detector. A CFAR detector is the signal process algorithm that controls the rate at which target detections are falsely declared. Given the current state of radar technology, CFAR algorithms are necessary elements of any automatic radar system. Unfortunately, CFAR systems are inherently vulnerable to degradation caused by large clutter edges multiple targets and jamming environments. This thesis presents eight popular and studied CFAR architectures. A comprehensive review of each system's structure, analysis and performance is detailed. Also the performance of each CFAR processor for two different inphase (1) and quadrature (Q) detectors: envelope approximation detector and the square law detector are compared numerically. In addition, each system is comprehensively compared to one another in the troublesome environments mention above. This thesis continues with the development of an original CFAR architecture, the excision greatest-of (EXGO). Although more complex, this processor is shown to be more robust than the other established techniques particularly in the presence of clutter edges, multiple targets, and electronic countermeasures (ECM) environments