RIS-assisted Data Collection and Wireless Power Transfer in Low-altitude Wireless Networks

2025-09-23

AI Summary

To address the challenges of energy constraints and poor channel conditions that hinder timely data collection from low-power Internet-of-Things devices (IoTDs) in remote areas of low-altitude wireless networks, this paper proposes a reconfigurable intelligent surface (RIS)-assisted unmanned aerial vehicle (UAV) joint optimization framework. The framework jointly optimizes RIS phase shifts, UAV 3D trajectory, wireless charging duration allocation, and binary IoTD scheduling to minimize both information age (AoI) and UAV energy consumption. We innovatively design the AO-IPDQN algorithm: alternating optimization reduces the complexity of high-dimensional RIS phase control, while an improved parameterized deep Q-network handles the hybrid action space. Simulation results demonstrate that the proposed method significantly outperforms baseline algorithms across diverse scenarios, achieving average reductions of 23.6% in AoI and 18.4% in UAV energy consumption—thereby enhancing data freshness and system energy efficiency.

Technology Category

• Intelligent Robots: Learning & Optimization for ROB
• Planning, Routing, and Scheduling: Optimization of Spatio-temporal Systems
• Search and Optimization: Algorithm Configuration

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Energy management for devices in mobile Web and WoT environments
• Responsible Web: Human-perceived consequences of algorithmic deployment on the web
• User Modeling, Personalization and Recommendation: On-Device user modeling, personalization, and recommendation

Abstract

Low-altitude wireless networks (LAWNs) have become effective solutions for collecting data from low-power Internet-of-Things devices (IoTDs) in remote areas with limited communication infrastructure. However, some outdoor IoTDs deployed in such areas face both energy constraints and low-channel quality challenges, making it challenging to ensure timely data collection from these IoTDs in LAWNs. In this work, we investigate a reconfigurable intelligent surface (RIS)-assisted uncrewed aerial vehicle (UAV)-enabled data collection and wireless power transfer system in LAWN. Specifically, IoTDs first harvest energy from a low-altitude UAV, and then upload their data to the UAV by applying the time division multiple access (TDMA) protocol, supported by an RIS to improve the channel quality. To maintain satisfactory data freshness of the IoTDs and save energy for an energy-constrained UAV, we aim to minimize the age of information (AoI) and energy consumption of the UAV by jointly optimizing the RIS phase shits, UAV trajectory, charging time allocation, and binary IoTD scheduling. We propose a deep reinforcement learning (DRL)-based approach, namely the alternating optimization-improved parameterized deep Q-network (AO-IPDQN). Specifically, considering that RIS typically contains a large number of reflecting elements, we first adopt an alternating optimization (AO) method to optimize the RIS phase shifts to reduce the dimension of the action space. Then, we propose the improved parameterized deep Q-network (IPDQN) method to deal with the hybrid action space. Simulation results indicate that AO-IPDQN approach achieves excellent performance relative to multiple comparison methods across various simulation scenarios.

Problem

• Optimizing data collection from energy-constrained IoT devices
• Minimizing UAV energy use and data age via joint optimization
• Developing a DRL approach for RIS-assisted UAV communication

Innovation

• RIS-assisted UAV for energy and data transfer
• Joint optimization of RIS phase shifts and UAV trajectory
• Deep reinforcement learning with alternating optimization