JoWUA

Abstract

Real-time activities on smart, ubiquitous devices have become a revolutionary way to improve health management, enhance safety, and automate daily life. This paper proposes a unified monitoring system that leverages wearable sensors, smartphones, and ambient IoT devices to gather and analyze data on human activity continuously. The system combines motion, physiological, and environmental cues to provide an in-depth view of user behavior in various and changing environments. The hybrid processing model is used: edge devices quickly identify activities with low latency, while long-term analysis, model optimization, and personal recommendations are performed on cloud resources. State-of-the-art machine learning algorithms are to be employed to categorize both straightforward and complex activities effectively, despite noise and abnormal user behaviour. The architecture is also focused on power-saving functionality and secure data transfer to ensure the device's long life and the user's privacy. The experimental findings indicate high precision, smooth scaling, and strong adaptability across a wide range of real-world applications, including healthcare tracking, smart homes, and workplace safety management. On the whole, this piece of work demonstrates the potential of ubiquitous smart devices to provide continuous, dependable, and context-aware activity monitoring that supports smart decision-making and improves the quality of life.