An Improved DV-maxHop Localization Algorithm for Wireless Sensor Networks

Abstract

Over the past few years, location information has contributed to a significant part in various Wireless Sensor Network (WSN) applications. Designing efficient localization algorithms is still a challenging task for many researchers. Our work emphasizes on range-free algorithm because of less cost and better performance in dense and indoor areas as compared to range-based algorithms. Mostly in real world scenario the network is anisotropic due to multiple holes and obstacles. Many recent range-free algorithms for anisotropic networks are complex with additional overhead of computation and assume even distribution of anchors, but it is not possible to deploy anchor nodes uniformly. DV-maxHop is one of the variants of DV-Hop algorithm which is an optimal range free algorithm for both isotropic and anisotropic networks, but still fails to achieve the desired accuracy in presence of non-uniform dissemination of anchor nodes. In context of exploring this limitation of DV-maxHop, we propose a novel strategy known as Improved DV-maxHop algorithm for localization. In our work, we correct the distances by modifying the average one hop of each link between anchor nodes and unknown nodes using a correction method. This correction helps the sensors to localize more accurately according to the distribution of sensor nodes in the network. Simulations has been performed and results indicate significant improvement is achieved with Improved DV-maxHop algorithm, when measuring metrics such as localization error, while varying other simulation parameters such as Anchor ratio and changing topology of anchors as well as sensors. Results prove that our proposed algorithm minimizes the localization error efficiently without requiring any additional hardware or increasing communication overhead.