Sign in Subscribe
Karlstad University

Machine Learning Research Offers Significant Energy Efficiency Gains in Cellular Internet of Things Devices

Strategic Insights

— 1 min read

Researchers at Karlstad University have published a new study on machine learning, proposing a framework that enables efficient energy use in Narrowband Internet of Things (NB-IoT) devices. The deployment of Cellular Internet of Things (CIoT) is expected to reach over six billion devices by 2030, with many located in remote areas where battery replacement or recharging would be difficult and expensive. To mitigate this challenge, the researchers developed the gradient-boosted learning optimization for battery efficiency (GLOBE) framework, which adjusts the radio layer of NB-IoT devices based on data transmission patterns and network conditions.

Key Takeaways:

  • The GLOBE framework reduces energy consumption by 30% to 75% compared to baseline configurations, offering significant benefits for both network operators and end devices.
  • The research aims to address the complex challenge of optimizing energy consumption in CIoT devices, considering their applications and operating environmental conditions.
  • The GLOBE framework is designed to enable swift and automated reconfiguration of NB-IoT devices, improving energy efficiency.
  • The study proposes a dynamic configuration of NB-IoT devices based on data transmission patterns and network conditions.
  • The research benefits both network operators and end devices by improving energy efficiency.
  • The study aims to contribute to the development of efficient energy use in CIoT devices, particularly in remote areas.

Statistics:

  • CIoT devices are expected to reach over six billion devices by 2030.
  • The GLOBE framework reduces energy consumption by 30% to 75% compared to baseline configurations.
  • The study proposes the GLOBE framework for dynamic configuration of NB-IoT devices.
  • The research aims to improve energy efficiency in CIoT devices, particularly in remote areas.

Sources:

  • Dynamic Nb-iot Configuration: a Machine-learning-driven Optimization Framework. Ieee Internet of Things Journal, 2025;12(19):40098-40114.
  • NewsRx. New Machine Learning Study Findings Recently Were Reported by Researchers at Karlstad University (Dynamic Nb-iot Configuration: a Machine-learning-driven Optimization Framework). Journal of Engineering. October 20, 2025; p 2061.

Read more