Joint Optimization of Data Collection for Multi-UAV-and-IRS-Assisted IoT in Urban Scenarios

Abstract

Due to their distinct economic efficiency and adaptability advantages, Unmanned Aerial Vehicles (UAVs) can serve as mobile data collectors, collecting data from Internet of Things Devices (IoTDs). As a promising emerging technology, the Intelligent Reflecting Surface (IRS) holds the potential to overcome architectural barriers and improve communication quality in urban environments. This study investigates the development of an IoT data collection system tailored for urban environments, leveraging the synergistic operation of multiple UAVs and IRSs. In light of the limited coverage capacity of an individual IRS, we deploy several IRSs, with multiple UAVs stationed at various base stations (BSs) to collect data from IoTDs. We propose a grouping genetic algorithm-independent double deep-Q network-alternating optimization (GGA-IDDQN-AO) approach, aiming to minimize the average mission completion time for a mission cycle. This approach optimizes both the deployment and mission allocation strategies of UAVs using the grouping genetic algorithm. Additionally, by integrating deep reinforcement learning with the alternating optimization algorithm, the flight trajectories of UAVs and IRSs’ phase shifts are fine-tuned. The effectiveness of the GGA-IDDQN-AO approach is validated through comprehensive simulations, which demonstrate that the integration of IRSs leads to a notable performance enhancement in the IoT data collection system.