Indoor localization and authentication using trilateration and speaker characteristics

• Main Author: Phyo, Ko Ko.
• Other Authors: Gan Woon Seng
• Format: Final Year Project
• Language: English
• Published: 2012
• Subjects: DRNTU::Engineering::Electrical and electronic engineering::Computer hardware, software and systems
• Online Access: http://hdl.handle.net/10356/49328
• Institution: Nanyang Technological University

The objective of this project is to research and develop the localization and authentication system, which includes Wi-Fi signal strength processing component for localization and Speech and Speaker recognition processing for authentication. The aim is to realize a new generation of access-control to facilities including hotels and apartments where knobs, card readers or sensors are no longer required. In this project, a desktop application developed in Visual Studio C++ IDE is used as the database server, a D-Link router is used as predetermined location beacon and a HTC android app developed in Eclipse platform with android SDK as end user human interface device for the system. The main purpose of the desktop application is, in the order, to set up wireless network connection with the android device, receive voice input from said device, perform speech recognition on the input, determine the phone (and hence the user) location using Wi-Fi signal strength and perform Speaker recognition on a list of users specified by the RSSI reading. It also hosts a simulation tool for stability and consistency testing, a user model training tool to train new user models and trilateration tool for network adapter selection and user database updating. The HTC android device serves as end user interface where the user can enter voice input to be used by the desktop application and display the result generated by the desktop application. The D-Link router serves as a calibrated static pre-determined location beacon.