Atmospheric Refractivity Tracking From Radar Clutter Using Kalman and Particle Filters

This paper addresses the problem of tracking the spatial and temporal lower atmospheric variations in marine and coastal environments. The method tracks the evolution of the range and height-dependent index of refraction using the sea clutter measured from sea-borne radars operating in the region. A split-step fast Fourier transform based parabolic equation approximation to the wave equation is used to compute the clutter return in complex environments with varying index of refraction. In addition, regional statistics are incorporated as prior densities, resulting in a highly nonlinear and non-Gaussian tracking problem. Various tracking algorithms such as the extended Kalman, unscented Kalman and particle Alters are used for tracking surface-based electromagnetic ducts frequently encountered in marine environments. Tracking performance of each Alter is also calculated and compared using the posterior Cramer-Rao lower bound. Even though the tracking performance of the Kalman Alters was limited by the high non-linearity of the parabolic equation, particle Alters proved to be very promising in tracking even the abruptly changing environments.