Accurate Analytical-Based Multi-Hop Localization With Low Energy Consumption for Irregular Networks

Abstract

In wireless networks, it is cost-effective to estimate the node-to-anchor distance by using the distribution characteristics of nodes. Unfortunately, in practice, the network topology may be irregular, which makes the estimated distances inaccurate. In this paper, by analyzing the distribution characteristics of nodes, we propose a novel scheme called Accurate Analytical-based Multi-hop Localization (AAML), which can greatly enhance estimation accuracy with requiring less communication and lower energy consumption. We first develop a new method that uses the common numbers of neighbors to estimate the distances between pairs of directly connected nodes during the network initialization phase. We then employ an optimal weighted matrix and hyperbolic estimation to reduce the effect of cumulative error. Finally, we utilize the weighted Taylor series to further improve the estimation accuracy; and utilize the geometric constraint to rectify the inaccurate estimated location, based on the relationship between the estimated locations and the anchor node locations. The proposed AAML method has been validated through analysis and experiments over different irregular networks and parameters, and AAML is demonstrated to be superior to existing methods in terms of estimated accuracy and energy consumption. The code is publicly available at https://pan.baidu.com/s/1k5o4aNGqa6iuxNskxRMMdw, pwd: n6ht.