Maximum a posteriori probability line tracking for nonstationary processes

The problem of estimating the instantaneous frequency (IF) of a discrete time FM signal in additive white noise is addressed. The IF laws are modeled as first-order stationary Markov processes defined on the unit circle and a recursive algorithm for determining the maximum a posteriori probability IF sequences is derived. The basic single signal algorithm is of complexity O(N³T) where T is the length of the signal segment and N is the number of subintervals of discretization of (0,2π) used for the maximization step. Some examples are given to illustrate the tracking ability of the method