Return to search

Application of Synthetic Aperture Radar with Wi-Fi for Indoor Localization

Indoor localization is the process of localizing people or objects inside a building in the same way GPS does in an outside environment. In recent years, researchers have successfully achieved improvement in indoor localization accuracy. Still there are many limitations to overcome in performing and achieving good accuracy in indoor localization. The interest in estimating the location of something inside a building with good accuracy is very strong. In this thesis we first propose an indoor localization technique relative to Wi-Fi access points along with a novel heuristic search based algorithm, named MuSLoc. Through simulation and comparative studies, we have shown that MuSLoc outperforms other indoor localization models without the help of fingerprinting or crowdsourcing about the environment. MuSLoc provides almost the same accuracy in LOS (Line of Sight) and NLOS (Non-Line of Sight) environments with regular infrastructure that has recently been provided by smart phones. This model doesn't require any additional hardware support in order to perform well.
Further, we propose another indoor localization based Wi-Fi device tracker model, named MSTracker, which is able to track both moving and non-moving devices inside a building. This model is also free from specialized infrastructure and can perform well without any training data information. Through real time simulation and analysis we have shown that it performs more accurately than other available models.
Through extensive simulations in a real time environment and analysis of performance comparatives with other available models, we have shown that both MuSLoc and MSTracker perform more accurately with COTS than any other method of indoor localization and tracking of objects inside a building. The complete package of MuSLoc and MSTracker can perform perfectly with recently available Wi-Fi modules and smartphones.

Identiferoai:union.ndltd.org:uottawa.ca/oai:ruor.uottawa.ca:10393/34467
Date January 2016
CreatorsNafi, Kawser Wazed
ContributorsNayak, Amiya
PublisherUniversité d'Ottawa / University of Ottawa
Source SetsUniversité d’Ottawa
LanguageEnglish
Detected LanguageEnglish
TypeThesis

Page generated in 0.0018 seconds