A deep learning approach to efficient drone mobility support

Abstract

The growing deployment of drones in a myriad of applications relies on seamless and reliable wireless connectivity for safe control and operation of drones. Cellular technology is a key enabler for providing essential wireless services to drones flying in the sky. Existing cellular networks targeting terrestrial usage can support the initial deployment of low-altitude drone users, but there are challenges such as mobility support. In this paper, we propose a novel handover framework for providing efficient mobility support and reliable wireless connectivity to drones served by a terrestrial cellular network. Using tools from deep reinforcement learning, we develop a deep Q-learning algorithm to dynamically optimize handover decisions to ensure robust connectivity for drone users. Simulation results show that the proposed framework significantly reduces the number of handovers at the expense of a small loss in signal strength relative to the baseline case where a drone always connect to a base station that provides the strongest received signal strength. CCS CONCEPTS • Networks → Mobile networks; Network mobility.

REQIBA: Regression and Deep Q-Learning for Intelligent UAV Cellular User to Base Station Association

“…This movement introduces the issue of UAV handovers, which complicates the association problem and requires an entirely new ML solution. Note that our work differs from existing works such as [24] and [25] in that we consider a throughput-maximisation problem for a UAV which communicates via a steerable directional antenna rather than an omni-directional one. This complicates the process of gathering environmental information for the association decision, which requires us to use a more complex ML solution to successfully optimise the UAV performance, as we will demonstrate in later sections.…”

“…The authors apply reinforcement learning to simultaneously select the serving BS and the allocated resource blocks with the aim of minimizing the uplink interference created by the UAV for the ground users, while keeping the rate of UAV handovers manageable. A similar problem is addressed in [25], where the authors intelligently select BS associations for a UAV moving along a known trajectory to minimise the rate of handovers. In [26], the The UAV chooses to associate with the BS at x 1 and centers its antenna main lobe on the BS location; the blue area W illuminated by the main lobe denotes the region where interferers may be found.…”

Toward Proactive Service Relocation for UAVs in MEC

“…However, if the application was relocated to D instead of C, the availability of the UAV application at MEC host D will be increased from two to five time slots. The same issue arise when relocating the application from E to F instead from E to G. This issue occurred because of very fast HOs from C to D and from F to G. Since UAVs are characterized by frequent and Pin-Pong HOs [11], [20], addressing this issue is mandatory to sustain an acceptable QoS and unlock the full potential of MEC for UAVs. Therefore, an optimized solution is required, which is the focus of this paper.…”