Training and classification of ballistic missiles using Hidden Markov model
Upendra Kumar Singh, Vineet Padmanabhan · 2013
This paper addresses the classification of different ranges of Ballistic Missiles (BM) for air defense applications using Hidden Markov Model (HMM). The classification is based on kinematic attributes like specific energy, acceleration, altitude and velocity which in-turn are acquired by radars. To meet the conflicting requirements of classifying short as well as long-range BM trajectories, we are proposing a formulation for partitioning the trajectory by using a moving window concept. This concept allows us to use parameters in localized frame which helps in reducing the problem of variety of trajectories to fit into the same model. Experimental results show that the Hidden Markov model is able to classify above 95 percentage within time of the order of milliseconds. To the best of our knowledge, this is the first time an attempt is made to classify ballistic missiles using HMM.