2020/02/04 by Md Moin Uddin Chowdhury, Walid Saad, Chowdhury, Md Moin Uddin +3
Engineering · #Advanced MIMO Systems Optimization #Advanced Wireless Communication Technologies #FOS: Electrical engineering #Signal Processing (eess.SP) #UAV Applications and Optimization #electronic engineering #information engineering
paper · pdf · doi:10.48550/arxiv.2002.01546
openalex publication_date 2020/02/04 · openalex created_date 2022/07/26 · openalex updated_date 2026/07/28
The pervasiveness of the wireless cellular network can be a key enabler for\nthe deployment of autonomous unmanned aerial vehicles (UAVs) in beyond visual\nline of sight scenarios without human control. However, traditional cellular\nnetworks are optimized for ground user equipment (GUE) such as smartphones\nwhich makes providing connectivity to flying UAVs very challenging. Moreover,\nensuring better connectivity to a moving cellular-connected UAV is notoriously\ndifficult due to the complex air-to-ground path loss model. In this paper, a\nnovel mechanism is proposed to ensure robust wireless connectivity and mobility\nsupport for cellular-connected UAVs by tuning the downtilt (DT) angles of all\nthe GBSs. By leveraging tools from reinforcement learning (RL), DT angles are\ndynamically adjusted by using a model-free RL algorithm. The goal is to provide\nefficient mobility support in the sky by maximizing the received signal quality\nat the UAV while also maintaining good throughput performance of the ground\nusers. Simulation results show that the proposed RL-based mobility management\n(MM) technique can reduce the number of handovers while maintaining the\nperformance goals, compared to the baseline MM scheme in which the network\nalways keeps the DT angle fixed.\n