An online target tracking protocol for vehicular Ad Hoc networks

Target tracking in vehicular ad hoc networks (VANETs) contributes to the design of many types of applications, namely: traffic management, security, car recovery and apprehension of an illegal runaway target. Inter-vehicular communication allows vehicles to participate and collaborate in the tracking process. For such applications, a large volume of data can be required to be transferred between the participating vehicles and a control center, which can easily congest the wireless network in a VANET and decrease the tracking efficiency if not managed properly. Therefore, one important challenge in this context is optimizing bandwidth usage to avoid collisions, delays and accelerate the overall tracking process. Thus, we propose a collaborative tracking protocol for VANETs based on a new strategy that we named virtual RSUs, which aims essentially to ensure the network communication coverage during the tracking process on the one hand, and on the other hand, to optimize bandwidth usage during the overall tracking process. In addition, in order to deal with uncertainties and enhance the tracking precision and further decrease the network load, we propose a theoretical pertinence level assignment strategy based on the Transferable Belief Model (TBM), that takes the target detection notifications as inputs. We believe this protocol holds potentials to serve as a basic algorithm to implement vehicle tracking applications for VANETs. Simulative study demonstrates clearly that the proposed protocol provides better performance in terms of network load for target tracking in a VANET as compared to a previous approach.