可重构智能表面和无人机辅助通信：深度强化学习方法

Reconfigurable intelligent surface and UAV-assisted communicationsA deep reinforcement learning approach Nguyen, Khoi Khac Award date:2022 Awarding institution:Queen's University Belfast Link to publication Terms of useAll those accessing thesis content in Queen’s University Belfast Research Portal are subject to the following terms and conditions of use • Copyright is subject to the Copyright, Designs and Patent Act 1988, or as modified by any successor legislation• Copyright and moral rights for thesis content are retained by the author and/or other copyright owners• A copy of a thesis may be downloaded for personal non-commercial research/study without the need for permission or charge• Distribution or reproduction of thesis content in any format is not permitted without the permission of the copyright holder• When citing this work, full bibliographic details should be supplied, including the author, title, awarding institution and date of thesis Take down policyA thesis can be removed from the Research Portal if there has been a breach of copyright, or a similarly robust reason. If you believe this document breaches copyright, or there is sufficient cause to take down, please contact us, citing details. Email:openaccess@qub.ac.uk Supplementary materialsWhere possible, we endeavour to provide supplementary materials to theses. This may include video, audio and other types of files. We endeavour to capture all content and upload as part of the Pure record for each thesis.Note, it may not be possible in all instances to convert analogue formats to usable digital formats for some supplementary materials. We exercise best efforts on our behalf and, in such instances, encourage the individual to consult the physical thesis for further information. Reconfigurable Intelligent Surface andUAV-assisted Communications: A DeepReinforcement Learning ApproachScott Johan Fischaber, BSc (Hon) Khoi Khac Nguyen A thesis submitted to theFaculty of Engineering and Physical Sciences in Queen’s University BelfastSchool of Electronics, Electrical Engineering and Computer ScienceQueen’s University Belfast A thesis submitted for the degree of Doctor of Philosophy October 2007April 4, 2022 Abstract Unmanned aerial vehicles (UAVs) and reconfigurable intelligent sur-face (RIS) have been considered as promising techniques for enhancingnetwork performance and coverage in wireless communication.TheUAV-assisted wireless networks are reliable, low-cost, and on-demandby using the agility and mobile features of the UAVs. The UAVs canprovide the maximum coverage and capacity for the targeted groundusers by adjusting their altitude. Their nimble mobility feature helpsthem avoid signal blockages and have better connections with theground users. However, due to the limitation of their on-board powerand flight time, it is challenging to obtain an optimal resource alloca-tion scheme for the UAV-assisted Internet of Things (IoT). The RISsreflect the signal from the transmitters to the receivers by controllingthe phase-shift value of a massive amount of scattering reflectors. Thereflected signals can be combined coherently to improve the receivedsignal or destructively to suppress the interference. In addition, thereliability and zero-delay are also notable advantages of the RIS insupporting reliable and low-cost wireless communications. Many of the devices used in IoT applications are energy-limited, andthus supplying energy while maintaining seamless connectivity for IoTdevices is of considerable importance.In this context, we propose a simultaneous wireless power transfer and information transmissionscheme for IoT devices with the support from RIS-aided UAV com-munications. In particular, IoT devices harvest energy from the UAVthrough wireless power transfer; and then, the UAV collects data fromthe IoT devices through information transmission. To characterise theagility of the UAV, we consider two scenarios: a hovering UAV anda mobile UAV. Aiming at maximising the total network sum-rate,we jointly optimise the trajectory of the UAV, the energy harvestingscheduling of IoT devices, and the phase-shift matrix of the RIS. We also investigate RIS-assisted multi-UAV networks that can utiliseboth advantages of UAVs’ agility and RIS’s reflection for enhancingthe network’s performance.Aiming at maximising the energy effi-ciency (EE) of the considered networks, we jointly optimise the powerallocation of the UAVs and the phase-shift matrix of the RIS. This thesis presents three major contributions.Firstly, we design anew UAV-assisted IoT system relying on the shortest flight path ofthe UAVs while maximising the amount of data collected from IoTdevices. Then, a deep reinforcement learning (DRL)-based techniqueis conceived for finding the optimal trajectory and throughput in aspecific coverage area.After training, the UAV has the ability toautonomously collect all the data from user nodes at a significant to-tal sum-rate improvement