Logo: to the web site of Uppsala University

Underground tunnels contain toxic, hazardous gases or lack oxygen. Therefore, providing the safety of personnel working in underground tunnels by monitoring oxygen, combustible or toxic gases is important. Wearable sensor networks are used in various applications such as oxygen monitoring. Some wearable sensor networks for gas monitoring are powered by batteries. Given the finite lifetime of batteries and the long term stay of underground personnel in these places, e.g. as a result of incidents, having a wearable sensor node with a stable power supply is crucial to conducting oxygen monitoring and notifying the emergency conditions. In order to provide a battery-less operation of the wearable sensor node, one approach is to use ambient energy sources. Considering the fact that the human body can survive only a few minutes without oxygen, the goal of our work is to investigate whether the power generated as a result of temperature difference between the body and ambient air is sufficient to power a sensor node for conducting oxygen detection every 20 seconds. Therefore, we present the development of a wearable sensor node for oxygen detection. The platform consists of a thermoelectric generator (TEG) converter, a sensing circuit for an electrochemical gas sensor and a power management circuit. Our experimental results with our platform indicate that a temperature difference of 6 degrees C between the body and the ambient air and storing the harvested energy in a capacitor ensure the autonomous operation of the wearable sensor node for measurements and alarm actuation conducted every 15 seconds. The period between the measurements could even become less than 15 seconds in underground tunnels as the temperature difference increases.

We present TaDA, a system architecture enabling efficient execution of Internet of Things (IoT) applications across multiple computing units, powered by ambient energy harvesting. Low-power microcontroller units (MCUs) are increasingly specialized; for example, custom designs feature hardware acceleration of neural network inference, next to designs providing energy-efficient input/output. As application requirements are growingly diverse, we argue that no single MCU can efficiently fulfill them. TaDA allows programmers to assign the execution of different slices of the application logic to the most efficient MCU for the job. We achieve this by decoupling task executions in time and space, using a special-purpose hardware interconnect we design, while providing persistent storage to cross periods of energy unavailability. We compare our prototype performance against the single most efficient computing unit for a given workload. We show that our prototype saves up to 96.7% energy per application round. Given the same energy budget, this yields up to a 68.7x throughput improvement.