MULTIPLE TARGET TRACKING AND MULTIPLE FREQUENCY LINE TRACKING USING HIDDEN MARKOV-MODELS

This paper considers the application of hidden Markov models (HMM's) to the problem of multitarget tracking, specifically, to the problem of tracking multiple frequency lines. The idea of a mixed track is introduced, a multitrack Viterbi algorithm is described, and a detailed analysis of the underlying Markov model is presented. Simulations show that in some cases, it is possible to avoid data association and directly compute the maximum a posteriori "mixed" track. Some practical aspects of the algorithm are discussed and simulation results presented.