Indoor Positioning System using Polarized Sweeping Light Beam

碩士 === 國立臺灣大學 === 資訊工程學研究所 === 104 === Benefit from the wide deployment of Light Emitting Diodes (LED), visible light communication technology has become an active research topic full of commercialization potentials. Taking advantage of this recent development, in this thesis, we proposes LiBeamScan...

Full description

Bibliographic Details
Main Authors: Yu-Lin Wei, 魏佑霖
Other Authors: Hsin-Mu Tsai
Format: Others
Language:en_US
Published: 2016
Online Access:http://ndltd.ncl.edu.tw/handle/11824117694781451612
Description
Summary:碩士 === 國立臺灣大學 === 資訊工程學研究所 === 104 === Benefit from the wide deployment of Light Emitting Diodes (LED), visible light communication technology has become an active research topic full of commercialization potentials. Taking advantage of this recent development, in this thesis, we proposes LiBeamScanner, an indoor positioning system that utilizes only a single custom light bulb and a light sensor to achieve extremely high positioning accuracy. With only a single light bulb as the transmitter, this system can perform extremely accurate positioning in environments without dense deployment of illuminaries. And the system only utilizes a simple light sensor in the receiver rather than a camera, and has low complexity and low energy consumption. This enables our design to be adopted in systems with more stingent constraints such as wearables. At the transmitting side, we borrow the idea of projectors to send information to corresponding location without interfering neighboring areas by controlling LCD pixels to generate narrow light beams, enabling fine-grained indoor positioning. Moreover, we propose a novel design utilizing dual lenses in the transmitter to modulate the information of stand-off distance, i.e. the height information, in the transmitted signals. Evaluation results of our prototype show that the error is less than 5 cm in most cases and the median error does not exceed 2 cm for the 2-dimensional position estimation, and 17 cm for the 3-dimensional position estimation.