Tutorial

Nan Xie and Henry Leung

Low Power Sensors and Machine Learning for Industrial IoT
It is believed that low power and ultra-low power sensors would outnumber any other IoT devices by 2030. LPWAN (Low Power, Wide Area Network) technology has stood out as a promising low cost, long-range solution that enables battery-powered or energy-harvesting sensors to provide multiple years of services. In this tutorial, we will provide a comprehensive overview of LPWAN and compare various technologies including LoRa, Sigfox, ZigBee, BLE, LTE-M, and NB-IoT etc. Industrial battery-powered sensor applications for smart cities and field study for underground sensor deployment will be illustrated. We will also walk through the end-to-end data integration steps from sensors, radio gateway, network server to the loud data platform using real life use case examples. Important security challenges and best practices for battery-powered sensors will be elaborated. Since low power sensors are constrained by power and resources, integration with computationally intensive Machine Learning (ML) for intelligent processing and decision making becomes a unique challenge. This tutorial will discuss and review various methods for applying ML to low power sensor solutions, including traditional centralized learning, federated ML, and TinyML for edge computing. Development trend and future research opportunities for low power sensors will also be presented.

IEEE websites place cookies on your device to give you the best user experience. By using our websites, you agree to the placement of these cookies. To learn more, read our Privacy Policy.