Utilization of Multi-Sink Architectures for Lifetime Maximization in Underwater Sensor Networks

In a typical underwater sensor networks (USNs) application, sensor nodes convey their collected data from the underwater environment to the central sink node. Improperly positioned central sink node causes two problems. First, sensor nodes consume high energy for transmission if the central sink node is located far away from the nodes where the network lifetime can be negatively affected. Second, long transmission links occurred in such a deployment case causes a USN to have high end-to-end delays since acoustic waves are typical carriers used for USNs. These problems can be mitigated by using multi-sink architectures where sensor nodes transmit their collected data to the nearest sink node/nodes. In this work, we propose a mixed integer programming (MIP) model that maximizes USNs lifetime and we investigate the impact of utilizing multi-sink architectures on network lifetime, end-to-end delay, and node energy consumption as compared to utilizing single-sink architectures. We show that single-sink USNs have at most 51% lower lifetimes than multiple-sink USNs. Furthermore, multiple-sink USNs reduce end-to-end delay and node energy consumption of single-sink USNs by at most 51% and 42%, respectively.