TY  - JOUR
T1  - CNN-LSTM Based Lonely Death Prevention Home AIoT System
AU  - Kim, Yong-Ho 
AU  - Oh, Sung-Hyun 
AU  - Kim, Jeong-Gon 
JO  - The Journal of Korean Institute of Communications and Information Sciences
PY  - 2025
DA  - 2025/1/1
DO  - 10.7840/kics.2025.50.7.1143
KW  - AI
KW  - Lonely Death
KW  - IoT
KW  - CNN-LSTM
KW  - Pattern Prediction
AB  - In According to recent statistics, 79.1% of solitary death cases occur among young and middle-aged adults, extending beyond the elderly population and affecting all age groups. Additionally, 78.8% of individuals at risk of solitary death are from single-person households. This highlights the growing need for technologies capable of continuously monitoring movements in single-person households and providing immediate response in emergency situations. This paper proposes a real-time monitoring system that tracks and predicts hourly movements in single-person households. To achieve this, we developed a solitary death prevention system utilizing motion data and AI (Artificial Intelligence) for time-series predictions. The system employs PIR sensors (Passive Infrared Sensor) strategically placed in multiple locations within the home to address blind spots and expand the detection range. By implementing a Home AIoT (Artificial Intelligence of Things) system using a CNN-LSTM (Convolutional Neural Network-Long Short-Term Memory) model, the practicality of the system is validated.