Robust Target Tracking with Discontinuous Signal for Passive Sensor Systems

This paper considers the tracking problem of a moving emitter, which discontinuously emits certain signals, in a passive sensor system. The aim is to estimate both the kinematic state of the emitter and the discontinuous property of the signal emission. To this end, we first build the signal emission model of emitter and the discontinuous measurement model of the passive sensor. Next, we formulate the tracking problem in the Bayesian estimation framework based on these models. We then propose a backward adaptive update time Kalman filter (BAUT-KF), which can update state estimation at arbitrary time, to solve this tracking problem. In order to better estimate the discontinuous property of the signal emission, BAUT-KF employs an adaptive strategy to adjust the state update time. We also propose a forward adaptive update time Kalman filter (FAUT-KF) which adopts a forward adaptive strategy based on BAUT-KF. The linear and Gaussian implementation of the proposed algorithms is also presented. Finally, both the simulation and real-world data from a passive sonar system demonstrate the robustness and effectiveness of the proposed methods.