Application Of RBF Neural Network And Support Vector Machines In Radar Target Recognition

With the development of modern radar, target recognition has attracted significant attention lately and will be playing an important role in the warfare of future. Based on machine learning theory, developing the practical and effective classifier is of great importance. Here in, the RBF Neural Net and Support Vector Machines are applied in radar target recognition. The software package is developed, consisting of many kinds of classifying algorithm.Firstly, according to the principle of RBF, a novel learning algorithm is proposed and results in better recognition performance, which is verified by the real measurement data of aircrafts.Secondly, on the basis of statistical learning theory, the support vector machines is used to classify different radar targets, which is also tested by real data. The results demonstrate the effectiveness of SVM.Lastly, the target recognition software is presented. Their generalization and robustness are discussed in details. The different algorithms are compared with each other by the IRIS data.