Wifi-Based indoor localization utilizing machine learning technique

Razali, Muhamad Akmal (2022) Wifi-Based indoor localization utilizing machine learning technique. Project Report. Melaka, Malaysia, Universiti Teknikal Malaysia Melaka. (Submitted)

	Text (24 Pages) Wifi-Based indoor localization utilizing machine learning technique.pdf - Submitted Version Download (2MB)
	Text (Full Text) Wifi-Based indoor localization utilizing machine learning technique.pdf - Submitted Version Restricted to Registered users only Download (8MB)

Abstract

Position estimation usmg global positioning systems (GPS) is not reliable in indoor environment due to weak signal penetration and the complex nature of indoor setting which causes severe signal attenuation. This project will explore the potential of utilising available wireless signal in indoor environment which is WiFi. We will implement fingerprinting technique to predict the position of mobile devices in a confined indoor space. The methodology involves two stages which are training stage and testing stage. In training stage, sufficient number of data will be measured in a study area. The data will be the received signal strength (RSS) measured by a mobile phone at predetermined reference points (RPs). The measured data will be trained using machine learning technique in Matlab platform to produce an indoor localization prediction model. In the testing stage, the produced model will be used to predict the location of mobile devices based on measured RSS. This project is expected to compare the accuracy produced by several machine learning techniques, and identify the best prediction model to be used in WiFi-based indoor localization.

Actions (login required)

View Item