MAP sequence estimation of microwave refractivity from radar clutter via a particle filtering implementation of the viterbi algorithm
This paper addresses the problem of predicting microwave propagation loss under ducting conditions by means of estimating the range and height varying index of refraction from observations of radar sea clutter returns. Specifically, the Fourier split-step solution to the parabolic equation for wave propagation is used to formulate the problem into a non-linear dynamic state-estimation framework. Solution for the maximum a posteriori (MAP) sequence estimate of the range-varying refractivity is achieved by extending notions of particle filtering framework to the Viterbi algorithm for state estimation. Real data results based on experiments performed off Wallops Island, Virginia are presented which quantify the proposed methods ability to predict propagation loss at 3 GHz.