Research On DV - Hop Based Location Technology In Wireless Sensor Networks

Wireless sensor network has good self organization, dynamic network topology and high reliability, while sensors have high integration, low power consumption, tiny size, which makes wireless sensor network have a broad application prospects. Meanwhile, localization is the most basic function as well as critical technology in wireless sensor network. Among all the localization algorithms, the DV-Hop algorithm is based on range-free scheme, which can avoid the error of direct ranging and finish positioning when the ratio of anchors is very low. In view of the literature, this paper focuses on the localization technology based on the DV-Hop algorithm in wireless sensor network, the main work is summarized as follows:(1) In order to analyze the performance of the original DV-Hop algorithm, the basic principle and implementation of DV-Hop algorithm are introduced, and the main sources of the positioning error are analyzed:the existence of misbehavior nodes, the hop distance error caused by network hollows, the estimation bias of average distance per hop and minimum hop counts, node calculation methods. Next we investigate four kinds of improved DV-Hop algorithms, and compare their positioning error and complexity through matlab simulations.(2) To solve the existed problems in the third stage of the original DV-Hop algorithm, a novel DV-Hop algorithm based on the reference anchor selection (RAS) is proposed. The improved steps are as follows:an unknown node first estimates its positions with different reference anchors, then the unknown node obtains the connectivity of each estimated node and compares the connectivity of the actual node with that of estimated nodes, eventually the estimated node with the minimum connectivity difference is selected as the final estimated position. The simulation results show that the RAS DV-Hop algorithm has a high positioning accuracy as well as good robustness when comparing with the original DV-Hop algorithm, difference DV-Hop (DDV-Hop) algorithm, anchor position re-estimation DV-Hop (APRE DV-Hop) algorithm and improved DV-Hop algorithm.(3) To avoid the influence of location error caused by network hollows in the original DV-Hop algorithm, an improved algorithm based on hollow anchors detection (HAD) is proposed, which corrects average distance per hop through detecting hollow anchors with appropriate threshold value. The simulation results show that the HAD algorithm has a significant improvement in location accuracy without increasing hardware overhead. Additionnally, the proposed algorithm can be combined with the RAS algorithm, and the location accuracy of the joint algorithm is higher than the RAS algorithm.