Authors: Abhishek Mondal, Deepak Mishra, Ganesh Prasad, George C. Alexandropoulos, Azzam Alnahari, Riku Jantti

Published on: February 05, 2024

Impact Score: 8.22

Arxiv code: Arxiv:2402.02957

Summary

• What is new: A novel approach leveraging multiple UAVs for data traffic offloading from terrestrial base stations, focusing on maximizing user association through optimizing UAV trajectories and user association indicators.
• Why this is important: The challenge of limited spectrum and coverage areas of terrestrial base stations failing to meet the escalating data demands of network users, particularly in IoT applications.
• What the research proposes: The utilization of unmanned aerial vehicles (UAVs) as flexible, mobile access points to offload data traffic from terrestrial networks, with a multi-agent reinforcement learning framework to optimize their trajectories and user associations.
• Results: Through extensive simulations, the proposed approach demonstrated superior average UAV association performance, outperforming benchmark techniques like Q learning and particle swarm optimization.

Technical Details

Technological frameworks used: Multi-agent reinforcement learning framework

Models used: Finite state Markov decision process, Distributed state action reward state action (SARSA) algorithm

Data used: nan

Potential Impact

Telecommunications providers, IoT service providers, UAV manufacturers and service providers

Want to implement this idea in a business?

We have generated a startup concept here: Skyflow Networks.