Radar target classification based on support vector machines and High Resolution Range Profiles

Abstract

In this study, the support vector machine (SVM) was used as a classifier to identify aerospace objects. Radar target identification based on high resolution range profiles (HRRPs) received much attention because of its reduced complexity than those using two-dimensional (2-D) ISAR images. Therefore range profiles were used as feature vectors to represent radar data. Data sets which are for training and testing were generated by using a program called radar target backscattering simulation (RTBS) for three different target types. The performance of the SVM was compared with other classification algorithms including statistical classification techniques such as maximum likelihood (ML) and fisher linear likelihood (FLL).