Optimizing Drone Deployment for Maximized User Connectivity in Areas of Interest Via Deep Reinforcement Learning

• Published in: Journal of network and systems management Vol. 33; no. 3; p. 49
• Main Authors: Rajashekar, Kolichala, Garg, Ashutosh, M. Baswade, Anand, Sidhanta, Subhajit
• Format: Journal Article
• Language: English
• Published: New York Springer US 01.07.2025; Springer Nature B.V
• Subjects: Bandwidths; Communications Engineering; Computer Communication Networks; Computer Science; Computer Systems Organization and Communication Networks; Connectivity; Deep learning; Disasters; Drone aircraft; Drones; Information Systems and Communication Service; Markov processes; Networks; Operations Research/Decision Theory; Optimization; Robotics; Unmanned aerial vehicles; Drone base station; Deep reinforcement learning; DBS deployment; Soft actor-critic (SAC) algorithm; Normalized continuous action space for reinforcement learning; Drone disaster emergency communications
• ISSN: 1064-7570; 1573-7705
• DOI: 10.1007/s10922-025-09924-1

In areas with limited communication capabilities, such as disaster zones, network service providers can deploy groups of Unmanned Aerial Vehicles (UAVs), commonly referred to as drones, serving as Drone Base Stations (DBSs) to temporarily supplement traditional communication infrastructure. In these regions, people who have survived the disaster adapt their positions to seek protection as the affected area gradually expands. Therefore, it is necessary to change the relative position of the DBSs to ensure adequate network performance and uninterrupted connectivity for users. This objective involves continuously updating the DBS position, which is subject to a long-term objective. To achieve this objective, we use Deep Reinforcement Learning (DRL) to adaptively modify the position of the DBS in response to the varying location of the User Equipment (UE). To this end, we design a Markov Decision Process (MDP) accounting for the continuous nature of the DBS position and the DBS 360-degree movement in the horizontal plane. This allows our solution to actively explore various positions of the DBSs, leading to significantly enhanced connectivity and bandwidth for the non-stationary user equipment (UEs). This work is the first to utilize action-space normalization in a continuous action space, using linear interpolation-based techniques commonly employed in robotics and related research fields. This approach allows for a comprehensive exploration of the continuous space, resulting in significantly enhanced optimization. We demonstrate that our approach can improve on previous research in this area while considering a much more complex optimization problem while providing uninterrupted connectivity to mobile UEs using multiple DBSs while maintaining stable bandwidth.