Sustainable and Low-AoI Cooperative Data Acquisition in UAV-Aided Sensor Networks
Xiuwen Fu, Mingyuan Ren
Abstract
Sustainable and low-latency data transmission is a major challenge in sparsely deployed sensor network scenarios. To overcome this challenge, we investigate unmanned aerial vehicle (UAV)-aided sensor network system, where the far-end UAV relays data to the near-end UAV, and then the near-end UAV transmits data to the base station (BS). For such a system, we consider the age of information (AoI) as a metric of data freshness and propose an adaptive UAV-aided sensor network cooperative data acquisition (AUCDA) scheme. First, we present a Gaussian mixture clustering algorithm based on ambiguous threshold to divide task areas for UAVs. Building upon this, we introduce a diffusion-based relay pairing method to establish relay relationships between UAVs. Finally, we propose a multilayer adaptive large neighborhood search (MALNS) algorithm to generate trajectories for data relaying between UAVs and trajectories for UAVs to return to the BS for charging. Extensive experimental results demonstrate that the AUCDA scheme can reduce the average system AoI. Furthermore, the AUCDA scheme enables each UAV to return to the BS for charging in a timely manner, thus significantly extending the system lifetime.