Three-dimensional Positioning. Distance Independent Of The Wireless Sensor Network Research

Wireless sensor network (WSN) is a sort of dynamic, distributed, self-organized network made up of lots of functional sensors with wireless communication abilities. They sense, collect and process the information of cover area of WSN, and publish the results to the observer. WSN could bring a great deal of reliable information anytime, anywhere and in any condition. WSN could be applied to national defense, national security, environmental monitoring, traffic control, medical treatment, manufacturing, anti-terrorists as well as anti-disaster. It could also be applied to all kinds of electronic products and embedded systems. In conclusion, WSN brings people many revolutionary effects to people's life and work. Localization technology is a foundation of wireless sensor network techniques.At present, the research on sensor network localization commonly on two-dimensional, seldom on three-dimensional, but in practical application, because limit to terrain (such as mountain area, hill, and ocean) or environment (such as different level of building or different height in air), the wireless sensor sometimes distribute in three-dimensional, not on plane, so only localization in two-dimensional is not enough, and expanding the research to three-dimensional is needed.This paper researches on three-dimensional localization. Three-dimensional localization has more errors in range and localization compute than two-dimensional localization, so how to control localization error is one of difficulty of three-dimensional algorithm. This paper advance rang-free three-dimensional localization algorithm based on traditional DV-Hop algorithm. The main trait of this algorithm is chose the best average network hop distance, furthest reduce localization error; improve the precision of localization algorithm.While the network has little anchor node or anchor node distribute odds, the unknown node can't localization due to no enough anchor node, this paper advance clustering-based three-dimensional localization algorithm against this situation. This algorithm divide the network into many clusters, unknown node relative localization, then incorporate neighborhood clusters, build global reference frame. When network has enough anchor nodes, it can make three-dimensional coordinate transform, and give unknown node absolute localization.