The Road towards the Internet of Batteryless Things

In this presentation we will dive into the need for Energy Harvesting, latest energy harvesting techniques and respective harvested energy information, Ultra-Low Power SoC architecture and features, SoC power consumption figures, BLE Channel Sounding and its advantages, and last but not the least the key battery-less IoT applications.

• On the battery problem in IoT, and the sustainability of batteries and batteryless devices
• What are the major challenges in computing and communications for batteryless IoT devices?
• An overview of the components of batteryless IoT devices
• A short overview of the main achievements of the IoBaLeT project
• Outlook and future work of the project consortium

The design experts from Quicksand and Sentineo first present a number of good design principles to realise reliable batteryless devices with a long lifetime. This will be followed by a short debate on the R&D innovations that are needed to accelerate the application of energy harvesting in ultra-low power (batteryless) devices.

The development of integrated circuits for energy harvesting presents a cutting-edge solution to powering low-energy devices in an increasingly connected world. This presentation explores the state-of-the-art advancements in energy-harvesting ICs, focusing on their ability to efficiently capture, convert, and store ambient energy. Key challenges include starting the system with very little energy, optimizing power conversion efficiency to and from the storage element. Real-world products are showcased to demonstrate practical implementations and the transformative potential of this technology across IoT and industrial applications.

• pros and cons of RF and light energy harvesting
• hybrid RF and light energy harvesting with communications capabilities

• Intro – problem statement: creating a repeatable test environment for quick iterations of energy harvesting setups
• Design of a novel emulation platform
• Hardware architecture
• Software models for harvesters, supercapacitors and power management systems
• Evaluation and results

• Demo 1 – imec-IDLab-UAntwerp:
  outdoor battery-free tracker
• Demo 2 – Atlas Copco:
  SmartClamp – energy harvesting from a power cable for machine monitoring
• Demo 3 – KU Leuven:
  cooperative light/RF-based localisation, in a setup with simultaneous wireless information and power transfer
• Demo 4 – imec-IDLab-UGent:
  battery-free 6TiSCH simulator demo
• Demo 5 – imec-IDLab-UGent:
  emulation platform for efficient development of sustainable batteryless sensor systems
• Demo 6 – imec-IDLab-UGent:
  wireless power transfer link for batteryless IoT
  
• Demo 7 – e-peas:
  power management solutions in a hybrid energy harvesting setup
  
• Demo 8 – VersaSense:
  ultra low power wireless networking in industrial environments
  

Sensor nodes that use batteryless energy harvesters offer an interesting alternative for long term deployment of sensors. However, these energy harvesters, equipped with supercapacitors instead of batteries, suffer from intermittent on-off behavior due to their limited energy storage capacity. This presentation discusses if such devices can still be part of industrial IoT networks such as WirelesHART, ISA100.11a or 6TiSCH.

• Overview of NB-IoT for low-power cellular connectivity
• Can NB-IoT be used for batteryless IoT devices?
• Overview of upcoming Ambient IoT 3GPP standard for energy harvesting devices
• Open issues and challenges with the current proposal

Industrial environments pose unique challenges for IoT networking, particularly when it comes to balancing energy efficiency, reliability, and scalability. This presentation explores how VersaSense leverages ultra-low-power wireless technologies to unlock new possibilities for monitoring previously unconnected industrial assets. Drawing on real-world examples from the manufacturing industry, insights and experiences will highlight how these solutions deliver specific monitoring intelligence while overcoming the constraints of harsh industrial settings.

Due to the unpredictable and dynamic harvesting environment, successfully running tinyML on the microcontroller of batteryless devices is still challenging. This presentation provides strategies for machine learning solutions to be compatible with energy-constrained IoT devices, and outlines future innovations that allow machine learning inference on constrained embedded devices.

For certain applications, a future of batteryless IoT solutions brings a tremendous potential value. This presentation will share some of the various design challenges and requirements from a wide range of use cases — from asset tracking, to condition monitoring and building optimization.