Information-Based Mobile Sensor Network Localization and Tracking of an Uncooperative Target

Abstract

This paper focuses on the localization and tracking of a radio frequency (RF) emitter with unknown waveform (uncooperative target) using an information-based centralized network of mobile, collaborative sensors (agents). Each agent utilizes the measured waveform parameters such as time difference of arrival (TDOA), frequency difference of arrival (FDOA), or differential received signal strength (DRSS). The proposed algorithm leverages Bayesian estimation to handle non-Gaussian measurements and it employs information-theoretic motion planning to move sensors to areas of high information-gain to minimize measurement uncertainty. Mutual information is computed for pairs of sensor through a four-permutation with replacement process. The generated mutual information surfaces are combined and then used to plan the motion of each sensor in the network. Simulation results involving TDOA measurements are presented to illustrate the effectiveness of the algorithm, where it is shown that three agents are able to accurately track a target that moves at the same speed as the agents.