LPWAN and Embedded Machine Learning as Enablers for the Next Generation of Wearable Devices

Area of Science:

• Computer Science
• Electrical Engineering
• Wearable Technology

Background:

• Wearable devices are increasingly integrated into daily life, primarily offering limited monitoring functions.
• Current wearables face hardware and power constraints, necessitating reliance on master devices like smartphones for data processing and cloud connectivity.
• The evolution of wearables is driven by advancements in communication and artificial intelligence (AI) technologies.

Purpose of the Study:

• To review and discuss the impact of low-power wide-area network (LPWAN) and tiny machine learning (TinyML) on wearable technology.
• To explore the implications and challenges associated with integrating LPWAN and TinyML into wearables.
• To experimentally analyze the long-range connectivity and embedded intelligence capabilities of an LPWAN-enabled wearable device.

Main Methods:

• Literature review of LPWAN and TinyML technologies for wearables.
• Experimental deployment of an LPWAN-integrated wearable device in a university campus setting.
• Analysis of the wearable device's connectivity range and the complexity of its embedded intelligence.

Main Results:

• Demonstration of enhanced long-range connectivity for wearable devices using LPWAN.
• Evaluation of the feasibility and complexity of embedding AI-driven intelligence within resource-constrained wearables.
• Validation of the significant potential of LPWAN and TinyML for next-generation wearable applications.

Conclusions:

• LPWAN and TinyML technologies are transformative for wearables, enabling advanced functionalities.
• Next-generation wearables will support a diverse range of novel services and applications due to these integrated technologies.
• The study highlights the practical benefits and challenges of implementing these cutting-edge paradigms in wearable systems.