Study Of Three-Dimensional Localization Approach Based On APIT In WSN

As the technologies of sensor, micro system, wireless communication and thecomputer developed, wireless sensor networks are widely applied into the areas ofdefense, environmental detection, logistics management and medical heath. Almost inall applications of WSN, location information of nodes is essential to the detection.However, the sensor nodes are deployed in the detecting area randomly and its locationis unable to be known. As a widely applied locationg system, Global Positioning System(GPS) can provide the relative high-precision of location, velocity and time information.But it can not satisfy the people due to its high overhead and terrible noise immunity.Furthermore, the two-dimensional localization can not satisfy the requirement of people,the three-dimensional localization becomes the urgently needed technology as therequirement of the WSN application improving. The research of this paper comes fromthe Provincial Natural Science Foundation. The study of this paper is on wireless sensornetwork nodes own spatial orientation based on APIT and refinement.This paper introduces a lot of typical localization algorithm which applied in WSNand analyzes the related works of the WSN localization including its developingdirections. It studys the three-dimensional localization technology based on APITcomprehensively. On this basis，a localization algorithm named IAPIT-3D (ImprovedThree-dimensional APIT) is proposed. This algorithm utilizes the “PerpendicularMedian Surface Cutting” to narrow down the targeted area, add two conditions toreduce the misjudgments. Then, it adopts the improved GRID SCAN to enhance therobustness to the instability of RSSI. At last, the paper propose a RSSI-weightedrefinement method according to some drawback that caused by the low-density of thebeacon nodes. The core idea of this paper is taking the reciprocal of distance betweennodes as the weight to decrease the error of localization.Through a mass of simulation, the localization and refinement reduce the“In-to-Out error”and“Out-to-In error”with almost no increment of the locating time.It improves the accuracy of the localization.