Superimposed HMM transient detection via target tracking ideas

The quickest detection of superimposed hidden Markov model (HMM) transient signals is addressed. It is assumed that a known HMM is always extant but a t an unknown time a second known HMM may also be present, and overlapped wi th the previous. Two approaches are proposed. The first treats the superimp osed HMMs as a unit with an expanded state space, thus converting the probl em of detecting superimposed HMMs into detection of a change in HMM, this b eing readily solved using a previously proposed procedure. Such an approach , though excellent in terms of performance, Is not suitable for the superpo sition of multiple HMMs with large state dimensions due to computational co mplexity. A second detection scheme (based on multiple target tracking idea s) with much lower computational needs but little loss in terms of performa nce, is therefore developed.