A Novel Adaptive Real-time Tracking Scheme for Underwater Networks

Abstract

Location information of mobile submersibles is critical to underwater networks. A considerable amount of researches have been carried out on non-real-time tracking to suffice many scenarios, while the lack of real-time tracking motivates more efforts to be addressed in Real-time underwater applications. In this paper a novel adaptive Real-time tracking scheme is presented, which can adaptively adjust the length of factor graph and then construct the resulting partial factor graph to track the positions of mobile submersibles. Since the proposed scheme tailors the basic factor graph into a partial factor graph with much shorter length in time, it makes the Real-time tracking feasible. Meanwhile, adaptively constructing the appropriate partial factor graph according to accuracy requirement of localization, can obtain high accuracy for localization and tracking. Simulation results using real sea-trail data show that our proposed scheme can adaptively trade off Real-time requirement and localization accuracy, achieving 12% to 61% improvement for tracking performance in the corresponding scenarios, when comparing with the classic TDoA (Time Difference of Arrival) localization method.