Development of a wireless sensor network for detection of human

• Main Author: Oo, Oo Khin
• Other Authors: Cheah Chien Chern
• Format: Final Year Project
• Language: English
• Published: 2017
• Subjects: DRNTU::Engineering::Electrical and electronic engineering
• Online Access: http://hdl.handle.net/10356/71733
• Institution: Nanyang Technological University

This report consists of the development of wireless sensor network for detection of human. With increasing population of elders in society, it is necessary to provide and develop new technologies that enhance the quality of life of elders. It is undeniable that fall injuries take high percentage of elderly accidents. Therefore, many fall detection solutions emerged to monitor the sudden fall of elders staying at home alone. The current solutions provide 70-80 percent of accuracy for fall detection. However, the user needs to wear a wearable device at all times. This leads to inconvenience for some users. The aim of this project is to develop a wireless sensor network that detects the presence of a human in an indoor environment by monitoring the changes of physical parameters without using any tracked devices attached to human. The project is broken down into 3 parts - to develop a wireless sensor network that is able to detect a human in the environment, to monitor and study the change in physical parameters in environment due to the presence of the entity and to analyse and provide methods to correctly detect the presence and fall of human in the environment. 8 wireless sensors were used in the experiment with 4 in one layer and the rest in another. Three different configurations were studied to select the suitable sensors’ arrangement for fall detection. Physical parameter Received Signal Strength indicator (RSSI) was used to observe the change in the environment. With results obtained from different configurations, the most suitable configuration was selected for fall detection. Furthermore, two data analysis approaches were proposed to accurately detect the fall of person. Experiments were conducted to verify the chosen configuration and data analysis approach.