Object Detector on Coastal Surveillance Radar Using Two-Dimensional Order-Statistic Constant-False Alarm Rate Algoritm

Abstract

This paper describes the development of radar object detection using two dimensional constant false alarm rate (2D-CFAR). Objective of this development is to minimize noise detection if compared with the previous algorithm that uses one dimensional constant false alarm rate (1D-CFAR) algorithm such as order-statistic (OS) CFAR, cell-averaging (CA) CFAR, AND logic (AND) CFAR and variability index (VI) CFAR where has been implemented on coastal surveillance radar. The optimum detection result in coastal surveillance radar testing when Pfa set to 1e-2, K^th set to 3/4*Nwindow and Guard Cell set to 0. Principle of 2D-CFAR algorithm is combining of two CFAR algorithms for each array data of azimuth and range. Order statistic (OS) CFAR algoritm is implemented on this 2D-CFAR by fusion rule of AND logic.The algorithm of 2D-CFAR is developed using Microsoft Visual C++ 2008 and the output of 2D-CFAR is plotted on PPI scope radar using GDI+ library. The result of 2D-CFAR development shows that 2D-CFAR can minimize noise detected if compared with 1D-CFAR with the same parameter of CFAR. Best performance of 2D-CFAR in object detection when Nwindow set to 128. The time of software processing of 2D-CFAR is about two times longer than the 1D-CFAR.