Optimized Range-Free Localization Scheme Using Autonomous Groups Particles Swarm Optimization for Anisotropic Wireless Sensor Networks

Abstract

Location information is a required concern in localization-based service application in the field of wireless sensor networks (WSNs). Distance Vector-Hop (DV-Hop) algorithm as the most typical rangefree localization scheme is much fitter for large-scaled WSNs. Its localization performance is good in even distributed networks. However, it demonstrated extremely poor accuracy under anisotropic networks. It is an urgent problem that need to be addressed. Accordingly, an optimized DV-Hop localization algorithm is put forward in this study with considering several anisotropic factors. Accumulated hop size error and collinearity are two main reason that led to low accuracy and poor stability. Hence, hop size error of anchors is reduced by introducing distance gap based on anchors. Besides, weighted least square method is adopted to replace the least square method to against anisotropic factors caused by irregular radio patterns. Moreover, an Autonomous Groups Particles Swarm Optimization (AGPSO) is employed to further optimize the obtained coordinate in the first round. It developed a novel method to determine localization coverage. The localization coverage is also added to be one evaluation metric in our study, which makes up for the lack of this evaluation indicator in most of the studies. Simulation results display good localization accuracy and strong stability under anisotropic networks. In addition, it also concluded that metaheuristic optimization algorithm and weighted least square method are more suitable to conquer anisotropic factor. It briefly points out a new direction for the future research work in the localization area under anisotropic networks.