A Time-Slotted Data Gathering Medium Access Control Protocol Using Q-Learning for Underwater Acoustic Sensor Networks

Contention-based medium access control (MAC) protocols for underwater acoustic sensor networks are designed to handle packet collisions that caused by long propagation delays. However, existing known suffer from relatively high collisions, which decrease system performance. To enhance performance, we propose a contention-based MAC protocol employs widely-popular machine learning technique, namely, Q-learning. Using Q-learning, the proposed allows nodes intelligently select back-off slots and accordingly schedule transmission of data packets such minimized at receiver. Unlike in protocols, not required exchange scheduling information, implies has low complexity overhead. Under varying traffic loads node numbers, is compared with state-of-the-art ALOHA-Q environment (UW-ALOHA-Q), multiple collision avoidance (MACA-U) exponential increase (EIED) protocols. Results demonstrate effectiveness terms energy efficiency, channel utilization, latency.