Research Of3-dimensional On-demand Indoor Localization System

Localization service is a popular application service applied to mobiledevices. As an emerging technology, localization is playing a significant rolein both commercial and residential and developing rapidly in recent decade.With the wide spread of social application based on smart devices,location-based service has gained a wide application. Therefore, there is agreater demand for localization service. The most common localizationservice system is Global Positioning System, GPS. By deploying24positioning satellites, GPS has achieved a good performance with an accuracyof10meters.Nevertheless, due to the block of walls and obstacles, GPS lacks a stableindoor signal and thus fails to make a good positioning. In order to do indoorlocalization, we need a set of new solution and system to make up the “Lastkilometer” of localization field. Similar to GPS System, Indoor LocalizationSystems also utilize wireless transmitters as anchors, receive wireless signals,and analyses the data to conduct localization.In this thesis, we present a3-dimensional on-demand IndoorLocalization System (3D-ODIL), which is fingerprint-free and can bedeployed rapidly in a multi-storey building. On basis of our design, we realizea prototype of3D-ODDL system. Firstly, we do the researches of mainstreamIndoor Localization Systems, and extract the core features. Secondly, weseparate our system into two parts, Horizontal Localization and Vertical Localization.On horizontal localization, we use Enhanced Field Division (EFD)algorithm, the main idea of which is to dynamically divide the field withindifferent signature area and position the target.On vertical direction, we propose Multi-storey Differential (MSD)algorithm. We establish a specific model for Buildings, and deploy Anchorson both ground floor and top floor. We use differential methods to calculatethe data, and distinguish the actual floor number. MSD Algorithms can greatlyreduce the number of anchors deployed.After we design the algorithms, we make a prototype for the3D-ODDLSystem. The system consists of four parts: Wireless Anchors, Mobile Device,Back-end Server and Front-end Display Application. Mobile Device collectswireless Signals and send the list to the Server; Server calculate and outputthe localization results to the Display Application.Finally, we do both simulations and experiments under differentconditions. The accuracy and performance have been validated through ourextensive analysis and systematic experiments. The errors on horizontaldirection are about4meters, and MSD works well.3D-ODDL presented in this thesis can solve some problems existed inindoor environment, which can do3-dimentional indoor localization inbuildings. The main idea and algorithms in this thesis provide a promisingway to solve indoor localization problems.