Comparison of Hough Transform and Particle Filter Methods of Emitter Geolocation using Fusion of TDOA Data

Abstract

This paper compares the performance of RF emitter geolocation algorithms based on the Hough transform and the particle filter. Three Hough transform methods are considered: (a) the generalized Hough transform, (b) the randomized Hough transform and (c) the hybrid Hough transform. In each case, the emitter is assumed to provide a signal from which time difference of arrival measurements can be made by pairs of mobile receiving platforms, such as fixed-wing UAVs or fast jets as well as rotorcraft. Typical emitters include cellphones and other types of communication equipment. The paper shows that the Hough transform outperforms the particle filter both in terms of the RMS positional error and the computational processing requirements.