ASL: An Accurate and Stable Localization algorithm for multi-hop irregular networks

Abstract

Accurate geographical information about nodes is essential in wireless multi-hop networks. Most existing localization algorithms focus on locating nodes in regular network environments, posing challenges for irregular multi-hop networks. To mitigate the impact of irregularities on localization, we propose an Accurate and Stable Localization algorithm (ASL). ASL first infers a hop threshold based on the distribution characteristics of anchors, eliminating erroneous distances and avoiding them in the localization process. Next, under the constraint of the hop threshold, each normal node constructs its sub-region, including it, based on the estimated distance to the anchors. These sub-regions can avoid the occurrence of unreliable localization results and assist in decreasing communication overhead. Finally, the SMA (Slime Mould Algorithm) with the Halton sequence is introduced to search for the optimally estimated locations of normal nodes, which tends to accelerate convergence and improve localization accuracy. Extensive simulations demonstrate that our proposed ASL outperforms state-of-the-art algorithms regarding accuracy and stability when facing network irregularities. Specifically, our proposed ASL achieves a median improvement in localization accuracy ranging from 16.96% to 83.66%.