应用局部保持投影法于无线网络室内定位 Wireless Local Area Network Indoor Positioning Using Locality Preserving Projects

Abstract

本研究利用无线网络接收到的讯号强度值进行室内定位，定位的算法我们引用了指纹特征比对法的概念，共分为离线阶段和在线阶段，在离线阶段我们应用局部保持投影法将接收到的讯号强度值投影到特征空间。在在线阶段，将接收到的实时讯号强度值利用最大似然法与离线阶段的资料作比对，估算出目前所在位置。实验结果显示，应用局部保持投影法可以将接收到的讯号强度值做前置处理，降低原始接收到的讯号维度，减少定位时的计算量，且在不影响定位准确率的同时，可以减少离线阶段所需事先收集讯号样本的时间，大幅节省事先收集讯号样本的工作量，从实验结果可知此算法应用于室内定位是可行的。 In this study, we utilize location fingerprinting approaches for indoor positioning by Wireless Local Area Network (WLAN) received signal strength. The approaches are divided into two parts including off-line and on-line stages. In the off-line stage, we collected received signal strength are projected onto the feature space of Locality Preserving Projects (LPP). In the on-line stage, we receive real-time signal strength compared with the off-line stage was collected data using Maximum Likelihood (ML) to estimate current location. Simulation results show that application of LPP not only reduce original dimension of the signal strength database but also reduce the computation. In addition, it can does not affect the positioning accuracy and reduce the off-line stage required prior to collection of signal samples at the same time. From Simulation results show that this algorithm is used in indoor location is feasible.