Energy-Throughput Trade-offs in a Wireless Sensor Network with Mobile Relay

In this paper we analyze the trade-offs between energy and throughput for links in a wireless sensor network. Our application of interest is one in which a number of low-powered sensors need to wirelessly communicate their measurements to a communications sink, or destination node, for communication to a central processor. We focus on one particular sensor source, and consider the case where the distance to the destination is beyond the peak power of the source. A relay node is required. Transmission energy of the sensor and the relay can be adjusted to minimize the total energy for a given throughput of the connection from sensor source to destination. We introduce a bounded random walk model for movement of the relay between the sensor and destination nodes, and characterize the total transmission energy and throughput performance using Markov steady state analysis. Based on the trade-offs between total energy and throughput we propose a new time-sharing protocol to exploit the movement of the relay to reduce the total energy. We demonstrate the effectiveness of time-sharing for minimizing the total energy consumption while achieving the throughput requirement. We then show that the timesharing scheme is more energy efficient than the popular sleep mode scheme.