An Adaptive Approach to Opportunistic Data Forwarding in Underwater Acoustic Sensor Networks

Reliable data transfer for underwater acoustic sensor networks (UASNs) is a major research challenge in applications such as pollution monitoring, oceanic data collection, and surveillance due to the long propagation delay and high error rate of the acoustic channel. To address this issue, an opportunistic data forwarding protocol [1] was proposed which achieves high packet delivery success ratio with less routing overhead and energy consumption by selecting the next hop forwarder among a set of candidates based on its link reliability and data transfer reachability. However, the protocol relies on fixed data hold time approach, i.e., each node holds data packets for a fixed amount of time before a forwarder discovery process is initiated. Depending on the value of the fixed hold time and deployment contextual scenario, this may incur large end-to-end delay. Moreover, lack of consideration of network condition in hold time limits its performance. In this paper, we propose an adaptive technique to improve its performance. The adaptive approach calculates data hold time at each node dynamically considering a number of 'node and network' metrics including current buffer occupancy, delay experienced by stored data packets, arrival and service rate, neighbors' data transmissions and reachability. Simulation results show that compared with fixed hold time approach, our adaptive technique reduces end-to-end delay significantly, achieves considerably higher data delivery and less energy consumption per successful packet delivery.