Joint sequential detection and estimation of Markov targets

The problem of joint detection and estimation when a variable number of noisy measurements can be taken is here considered in the case that the signal to be detected is generated by a dynamic system with a Markov evolution and the parameter to be estimated is the trajectory of the state evolution of the system itself and/or it final state (position). Starting from previous sequential rules, different sequential strategies are proposed and assessed: they are aimed at maximizing the performance of either the detection or the track estimation or the position estimation. Bounds on the performances of the proposed procedures in terms of the system parameters are derived and computational complexity is examined. Also, numerical experiments are provided to elicit the interplay between parameters and system performances and to quantify the gain with respect to other fixed-sample-size procedures.