Comparison of support vector machines and neural networks in an electronic attack application

In this paper, we consider flare dispensing which is one of the cost-effective electronic attack techniques widely used by air platforms to protect themselves from heat seeking infrared missiles (IRGM-Infrared Guided Missile), and compare classification of successful and unsuccessful flare dispensing programs against a chosen missile seeker via support vector machines (SVM) and artificial neural network (ANN). In this work, the engagement between an IRGM with a seeker using pulse width modulation (PWM) and an air platform which tries to escape from this threat by dispensing flare is simulated. The results show that SVM performs better than ANN in classifying successful and unsuccessful flare dispensing programs.