Application research on optimizing TDMA scheduling strategies in tactical data links using deep reinforcement learning
Teh‐Lu Liao, Chia-Hung Kao · IET conference proceedings. · 2025
Time Division Multiple Access (TDMA) scheduling strategy optimization based on Deep Reinforcement Learning (DRL) for Tactical Digital Information Links (TADIL). Unlike traditional TDMA methods, this investigation focuses on slot allocation adjustment in Mobile Ad-hoc Networks (MANETs) to enhance transmission efficiency and adapt to network dynamics. Through operation cases in radio networks, the proposed method demonstrates its effectiveness in significantly reducing communication delays for high-priority tasks while providing stable data transmission channels for low-priority tasks. Compared to existing methods, our TDMA scheduling strategy shows better adaptability to dynamic data rates while ensuring Quality of Service (QoS).