期刊
IEEE TRANSACTIONS ON VEHICULAR TECHNOLOGY
卷 71, 期 7, 页码 7691-7702出版社
IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC
DOI: 10.1109/TVT.2022.3163430
关键词
UAV; vehicular networks; spectrum sharing; resource allocation; reinforcement learning
资金
- National Key Research and Development Program of China [2018YFE0206800, 62025105]
- Chongqing Municipal Education Commission [CXQT21019]
- Natural Science Foundation of Chongqing [cstc2020jcyj-msxmX0918]
Vehicular networks play a crucial role in the era of autonomous driving, with UAVs showing potential to assist in service provision while facing challenges of energy consumption and QoS. Joint optimization of factors such as content placement and spectrum allocation can enhance UAVs' energy efficiency and meet user demands.
Vehicular networks are envisioned to deliver data transmission services ubiquitously, especially in the upcoming autonomous driving era. Accordingly, the high data traffic load poses a heavy burden to the terrestrial network infrastructure. Unmanned Aerial Vehicles (UAVs) show enormous potential to assist vehicular networks in providing services. In a dense UAV-assisted vehicular network with a large number of users, spectrum sharing is leveraged for alleviating the spectrum scarcity. However, the increasing data traffic still leads to the UAV energy consumption problem. This paper considers a specific network scenario where a UAV transmits its cached content files to vehicular users over UAV-to-vehicle (U2V) links while vehicle-to-vehicle (V2V) links reuse the U2V spectrum for safety-critical message exchanges. To improve the UAV's energy efficiency while guaranteeing users quality of service (QoS), we jointly optimize content placement, spectrum allocation, co-channel link pairing, and power control, which are the key factors affecting energy efficiency and QoS. The joint optimization problem is formulated as a mixed-integer nonlinear programming (MINLP) problem, which is solved by combining Hungarian and DDQN. We perform system performance evaluations, demonstrating that our approach can not only improve the UAV's energy efficiency while satisfying the users QoS requirements but also increase the timeliness of making decisions.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据