Channel Learning-Based Relay Selection in Underwater Acoustic Sensor Network

This paper proposes a channel learning-based relay selection method of cooperative communications to improve packet delivery ratio and reduce latency under the low propagation channel characteristics in underwater acoustic sensor networks. The proposed method estimates the time-varying channel using the sound speed profile based estimator to calculate the link capacity and transmits data packets to the selected best relay and candidate relays based on link capacity. Simulation results show that the proposed method performs better packet delivery ratio, latency, and throughput than the conventional SNR feedback and random selection methods.