AoI-Based Sensor Selection for Target Tracking in Asynchronous Sensor Networks

Due to communication delays and other limitations of wireless sensor networks (WSNs), asynchronous sensor selection is necessary for target tracking. However, not much work has been formulated for this problem. Inspired by the concept of data freshness, this article applies age of information (AoI) to sensor selection for target tracking in the presence of random communication delays. In this regard, we formulate several AoI-based selection designs to measure the value of asynchronous measurements. The first formulation minimizes the time-average AoI to achieve enhanced performance, the second formulation attempts to set an AoI deadline constraint, and the third formulation penalizes the updated delay using an AoI-based penalty function. These AoI-based formulations are applied to target tracking by combining AoI with mutual information (MI) for asynchronous sensor selection. The proposed formulations are then solved by the nondominated sorting genetic algorithm (NSGA-II) and the greedy approach. Finally, the selected sensors are fused by the asynchronous fusion approach. Simulation results validate the effectiveness of the proposed asynchronous sensor selection methods compared with traditional approaches.