Development of a device-free passive tracking system for home based rehabilitation

• Main Author: Ong, Kang Yu
• Other Authors: Cheah Chien Chern
• Format: Final Year Project
• Language: English
• Published: 2018
• Subjects: DRNTU::Engineering
• Online Access: http://hdl.handle.net/10356/74874
• Institution: Nanyang Technological University

Singapore’s ageing population and small healthcare labour force fostered an interest in healthcare technology to ease the increasing workload faced by the healthcare industry. There have been many initiatives by the government’s smart nation program. However, those initiatives are mostly active system, where the patient must wear a device. Hence, result in issue such as unwillingness and forgetfulness on the patient’s end. To address this problem, this paper aims to develop a real-time device-free indoor positioning system specifically to address the stroke patient rehabilitation process. The method employed is a radio frequency based probabilistic fingerprinting technique. Firstly, constructing the radio map using the system’s RSSI on selected locations. Next, using a Kalman filter to smoothen the raw RSSI of the system. Next, using the probabilistic model according to Bayes’ theorem to locate a match between the real-time RSSI and the fingerprinted radio map. Therefore, returning the pre-defined coordinate based on the matches in the probabilistic model as patient’s location. Due to the noisy environment, the stability of the system is affected. Hence, two stabilising algorithms, thresholding and weighted neighbour algorithm are used to stable the system. The result indicate that the estimated position has a 0.8-meter mean position error and a 1.52m mean distance error.