http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
Reinforcement Learning Based Optimization for Drone Mobility in 5G and Beyond Ultra-Dense Networks
Jawad T.(자와드),Kim, Ajung(김아정) 한국통신학회 2021 한국통신학회 학술대회논문집 Vol.2021 No.2
Drone applications in 5th generation (5G) networks mainly focus on services and use cases such as providing connectivity during crowded events, unmanned aerial vehicle (UAV) traffic management, and internet of things in the sky. 4G and 5G cellular networks face various challenges to ensure dynamic control and safety of drone mobility to deliver these enhanced services. The baseline greedy handover algorithm only ensures the strongest connection between the drone and small cells, so the drone may experience several handovers. Intended for fast environment learning, the machine learning technique such as Q-learning helps the drone to fly with minimum handover cost along with robust connectivity. In this paper, we propose a Q-learning based approach evaluated in three different scenarios. Simulation results demonstrate that the proposed algorithm can effectively minimize the handover cost in a learning environment.
Drone Mobility Optimization in 5G and Beyond Ultra-Dense Networks using SARSA Reinforcement Learning
Jawad T(자와드),Kim, Ajung(김아정) 한국통신학회 2021 한국통신학회 학술대회논문집 Vol.2021 No.6
Drone applications in 5th generation (5G) networks mainly focus on services and use cases such as providing connectivity during crowded events, unmanned aerial vehicle (UAV) traffic management, and internet of things in the sky. 4G and 5G cellular networks face various challenges to ensure dynamic control and safety of drone mobility to deliver these enhanced services. The baseline greedy handover algorithm only ensures the strongest connection between the drone and small cells, so the drone may experience several handovers. Intended for fast environment learning, the machine learning technique such as Q-learning (SARSA) helps the drone to fly with minimum handover cost along with robust connectivity. In this paper, we propose a Q-learning based approach evaluated in three different scenarios. Simulation results demonstrate that the proposed algorithm can effectively minimize the handover cost in a learning environment.