With an aging global population, it is increasingly a matter of concern of how to keep older individuals remaining in their homes safe and healthy. Invasive or disruptive monitoring methods such as cameras or wearables are undesirable. A more privacy-conscious and less obtrusive means is needed to monitor movement patterns and detect deviation from the norm, which could be an indicator of health risks, including falling and mobility loss.

Sensor Installation:

• LoRaWAN motion sensors will be placed in different rooms of the house.
• The sensors will record movement data for a predetermined period (e.g., several weeks).

Data Collection:

• The sensors will transmit motion event, timestamp, and location information to a central database.
• The data will be collected and pre-processed for analysis.

Machine Learning-Based Data Analysis:

• A Python-based ML algorithm will process movement patterns.
• The model will establish a baseline of normal daily movement patterns.
• Exceptions (e.g., no movement for extended periods or unusual movement patterns) will be flagged for investigation.

The aim of this project is to test the implementation of Tektelic's cost-effective LoRaWAN sensors at monitoring the everyday movement of an elderly person (my grandmother) at home. By feeding collected data through the application of Python's machine learning (ML) algorithm, the study attempts to determine patterns and relationships in her movement. The goal is to check if abnormalities in predetermined patterns may serve as symptoms of potential ailments.

Expected Data:

• Movement logs by room with timestamps.
• Frequency and duration of movement over the course of a day.
• Changes in movement between days of the week.

Analysis:

• The ML algorithm will cluster similar patterns and identify trends.
• A deviation-detection mechanism will be implemented to flag anomalous patterns.
• Correlation between movement patterns and external factors such as time of day, sleep cycles, or room preference can occur.

This project will demonstrate the feasibility of using LoRaWAN sensors for non-intrusive monitoring of the elderly. If successful, it would be a cost-effective and privacy-friendly alternative to existing monitoring systems. The insights from movement pattern analysis can help caregivers detect early indications of health decline or mobility issues, which can improve care for the elderly.

Tural E, Lu D, Austin Cole D. Safely and Actively Aging in Place: Older Adults’ Attitudes and Intentions Toward Smart Home Technologies. Gerontology and Geriatric Medicine. 2021;7. doi:10.1177/23337214211017340

Cicirelli G, Marani R, Petitti A, Milella A, D’Orazio T. Ambient Assisted Living: A Review of Technologies, Methodologies and Future Perspectives for Healthy Aging of Population. Sensors. 2021; 21(10):3549. https://doi.org/10.3390/s21103549

Stavropoulos TG, Papastergiou A, Mpaltadoros L, Nikolopoulos S, Kompatsiaris I. IoT Wearable Sensors and Devices in Elderly Care: A Literature Review. Sensors. 2020; 20(10):2826. https://doi.org/10.3390/s20102826

I would like to express my deepest gratitude to my grandmother for giving me permission to conduct this research and for showing patience and understanding throughout the process. I also highly appreciate the advice and facilities provided by Nurana Verdiyeva, whose professional input has played a key role in the development of this research.

Massive thank you to Tektelic for access to their low-cost LoRaWAN sensors so I could experiment with novel movement tracking solutions. Thank you to my school and Calgary Youth Science Fair for allowing me the opportunity to present my findings.

Lastly, I would like to thank my friends and family for their encouragement and to everyone who has contributed in any way towards the success of this project.