Fairness-Guided DQN-Based TXOP Control for NR-U/Wi-Fi Coexistence

The paper proposes a fairness-guided Deep Q-Network (DQN) to adapt NR-U transmission opportunities (TXOP) in unlicensed spectrum coexistence with Wi-Fi. It uses two reward designs, Q1 for strict per-node throughput balance and Q2 for broader fairness, showing that the system can be configured to prioritize different objectives like strict fairness or utility improvement.

Open the full episode

More from AI Radio