Date of Award

8-2019

Document Type

Campus Access Thesis

Degree Name

Master of Science (MS)

Department

Computer Science

First Advisor

Wei Ding

Second Advisor

Ping Chen

Third Advisor

Dan Simovici

Abstract

There has been a lot of research in the field of outdoor navigation. With popular technologies like GPS being used in smartphones and other devices, finding your current location and the route to your destination in outdoor areas has become very easy. As GPS signals are not accessible in indoor areas for many reasons like poor reception of satellite signal in indoor locations, these technologies cannot be used for Indoor Navigation. So, there has been research on other techniques and technologies for Indoor Navigation. Some of the techniques include Bluetooth beacon positioning, Wi-Fi based positioning and using hardware-based techniques which involve using hardware like cameras, RFID, infrared, ultrasound, and laser devices. My research involves using Wi-Fi data, specifically the MAC ID and strength of the signal to apply machine learning techniques, especially One-Shot Learning for classification using the chosen Wi-Fi features for Indoor Navigation. One of the machine learning techniques investigated is K-Nearest Neighbor algorithm where we use the MAC ID and strength of the access points to classify positions in the indoor area. Another direction I investigated is One Shot Learning with the help of Memory Augmented Neural Networks. This technique has the dual ability of using techniques to learn a method slowly over time that can be used to learn useful representations of data and use an external memory to quickly learn data that it has not seen before, after showing it the data only once. With real-world data I conducted experiments to compare the performance of the K-Nearest Neighbor algorithm and the One Shot Learning technique.

Comments

Free and open access to this Campus Access Thesis is made available to the UMass Boston community by ScholarWorks at UMass Boston. Those not on campus and those without a UMass Boston campus username and password may gain access to this thesis through resources like Proquest Dissertations & Theses Global or through Interlibrary Loan. If you have a UMass Boston campus username and password and would like to download this work from off-campus, click on the "Off-Campus UMass Boston Users" link above.

Download
Off-Campus UMass Boston Users

Share

COinS