Unmanned Aerial Vehicle Assisted Timely Data Collection in IoT Networks with Faulty Sensors

Thanks to their low deployment costs and high flexibility, unmanned aerial vehicles (UAVs) have been widely applied for data collection in Internet of Things (IoT) networks. However, the unpredictable status updates from deployed sensor nodes (SN) and potential failures of SN can affect the freshness of information. In this paper, we investigate the UAV data collection in monitored areas with redundant SNs to alleviate the impact of potential SN failures. The SNs generate status updates at random time slots, which will be collected by UAVs and transmitted to a data center. In order to minimize the average age of synchronization (AoS) of the network and the UAV energy consumption, we formulate a Markov decision process and adopt dueling double deep Q network (D3QN) method to optimize UAV flight trajectories. Simulation results demonstrate that our method achieves better performance compared with baseline policies in environments where SN failures occur.