History-based Multi-node Collaborative Localization In Mobile Wireless Ad Hoc Networks

Recent years have witnessed a growing interest in localization algorithms for wireless ad hoc networks. localization technology has been widely application used in the fields of military, industrial and civilian, for example, battlefield collaboration and layout, the positiondetermined between staffin the process of disaster emergency,localization and tracking of vehicles in application of intelligent transportation,etc.However, the localization of wireless ad hoc network faces with more and more challenges, for example, one of challenges is how to ensure the positioning precision while prolonging the network life time.Currently, most existinglocalization algorithms, focus on the localization for dense ad hoc networks, which improves the localization accuracy byincreasing the density of anchor nodes. However, there is little work on the localization for sparse hoc networks.In this paper, based on the number of reference nodes, we propose a distributed localization algorithm, i.e., history based multi-node collaborative localization algorithm(HMCL), which provides a potential approach for localization in sparse ad hoc wireless networks. In the proposed HMCL algorithm, we exploit a new motion model to filter the imprecise estimation values based on the historical position information of nodes, which can improve the localization accuracy and reduce the computation overhead and energy consumption. Moreover, we utilize different strategies to achieve the localization of nodes with different priorities measured by the distance information between neighbor nodes. We verify through experiment that the proposed algorithm provides better performance in terms of localization precision andenergy consumption. Besides, we also analyze the effect of the number of neighbor nodes, node density and moving speed of nodes on localization precision.