Frequency-Wavenumber Tracking Using Hidden Markov Models

Abstract

This correspondence extends earlier work by the authors on multiple frequency line tracking using hidden Markov models (HMM's) to also include wavenumber estimation. The main idea is to model the frequency and wavenumber of each target using HMM and then to track these signals using a Viterbi algorithm. The input measurements to the Viterbi algorithm are two-dimensional (2-D) Fourier transforms of the array output signals. Several supporting simulations show that the approach works well, although it is computationally expensive. © 1993 IEEE