A Copy-At Retransmission Strategy for Improved Wireless Sensor Network Lifetime and Reliability

Abstract

A Wireless Sensor Network (WSN) is composed of tiny autonomous sensors with limited battery power. WSNs are employed to observe specific fields of interest. In this paper, we study the energy efficient and reliable network design problem using a mathematical programming framework. Energy efficiency is vital since battery replenishment is not always viable and the network lifetime is measured as the time until the first sensor exhausts its energy. Moreover, reliability is important since sensors are mostly deployed unattended and transmission of data fully and correctly is obviously critical. We develop a retransmission strategy originated from the Pareto principle and the scale-free property of complex networks. In our modified hop-by-hop reliability definition, sensors forwarding data directly to the central node must perform retransmission. Central node is the sensor with the highest data transmission load and our motivation is to secure the transmission of data passing through the central node against malicious attacks or technical failures. To this end, we present a mixed 0–1 integer programming model and an efficient heuristic. Our test results show an improvement of 80.5% in network lifetime and of 86.3% in redundant data overhead when compared with the classical conservative data redundancy approaches. We provide extensive test results, which reveal the contribution of our strategy in several other strategic design dimensions.