Skip to Main Content
Geographic routing is an attractive option for large scale wireless sensor networks (WSNs) because of its low overhead and energy expenditure, but is inefficient in realistic localization conditions. Positioning systems are inevitably imprecise because of inexact range measurements and location errors lead to poor performance of geographic routing in terms of packet delivery ratio (PDR) and energy efficiency. This paper proposes a novel, low-complexity, error-resilient geographic routing method, named conditioned mean square error ratio (CMSER) routing, intended to efficiently make use of existing network information and to successfully route packets when localization is inaccurate. Next hop selection is based on the largest distance to destination (minimizing the number of forwarding hops) and on the smallest estimated error figure associated with the measured neighbor coordinates. It is found that CMSER outperforms other basic greedy forwarding techniques employed by algorithms such as most forward within range (MFR), maximum expectation progress (MEP) and least expected distance (LED). Simulation results show that the throughput for CMSER is higher than for other methods, additionally it also reduces the energy wasted on lost packets by keeping their routing paths short.