Child activities monitoring and management system using Smart wearables

• Main Author: Teo, Zhi Yun
• Other Authors: Oh Hong Lye
• Format: Final Year Project
• Language: English
• Published: 2017
• Subjects: DRNTU::Engineering::Computer science and engineering
• Online Access: http://hdl.handle.net/10356/70226
• Institution: Nanyang Technological University

Improvements in technology have made wearable technology smaller and have more capabilities while lowering costs. Such improvements contributed to the increasing popularity of wearable devices as they are more practical and functional for users. The devices carry an array of sensors and hardware that allow the device to have multiple functions such as health monitoring, controlling other devices, and improving convenience. In the previous project, a collection of smart watches was explored as a possible means to monitor child activities in the classroom and excursion setting. The smart watch will assist the teacher/ care-taker by keeping track of the students/children and alert him/her when the child runs out of range. This enables the teacher to perform other tasks without having to constantly keep an eye on all the children, which could be difficult if the class is big. The project was implemented but had limitations due to hardware restrictions of the developmental kit. This project is an extension of the previous project with the purpose of overcoming those limitations and adding additional features. There are three parts in this project; Firstly, improving battery life of the central node (teacher’s device) through duty cycling to reduce power consumption of the radio frequency module. Secondly, overcoming the limitations of a star topology by allowing each end device to act as an intermediary node. And lastly, identifying and implementing a fall detection algorithm that will warn the teacher when a child falls down. The improvements were implemented over a period of eight months as three separate applications. Duty cycling reduced battery consumption by more than 50%. The star topology network was successfully extended by forwarding messages from a device out of range to the central node through a device within the range of both devices. Two fall detection algorithms were derived and tested based on readings from simulated falls and daily activities. Node to node tracking and fall detection were successfully integrated. Measurements and tests were conducted to ensure the improvements are working as planned, fulfilling the project requirement.