Low power sensor nodes

Signal processing for sensor nodes and wearables: driving Big Data and the Internet of Things

Very low power consumption sensor nodes are at the heart of the emerging Internet of Things, and the huge quantities of data driving the Big Data revolution. We make ultra low power consumption signal processing for embedding in sensor nodes to enhance their functionality and battery life. We focus on the red blocks in the diagram above, providing real-time analysis and feedback, and the required PC based and offline analyses to support this.

Our work spans:

  • Signal processing design in Matlab
  • PCB design of sensor nodes
  • Fully custom microchip design for ultra low power processing
  • Energy harvester powered sensor nodes

Our approach

Our focus is on wearable sensor nodes for human monitoring, complimenting our work on novel neurotechnology. We create electronics for monitoring a range of physiological parameters, such as heart rate, and specialise in creating onboard signal processing for smart sensors. As one example, we have demonstrated how the Continuous Wavelet Transform can analyse signals from the heart in real-time using only 1.4 nW of power

Benefits of onboard signal processing

  1. Reduced system power consumption
  2. Increased device functionality
  3. Reliable and robust operation despite unreliable wireless links
  4. Minimized system latency
  5. Reduction in the amount of data to be analysed offline
  6. Enabling of closed-loop recording and stimulation devices
  7. Better quality recordings (e.g. with motion artefact removal)
  8. Real-time data redaction for privacy

Technologies

Our technologies are underpinned by designing signal processing algorithms, and hardware for implementing algorithms, which can operate in the sensor node while consuming little power. 

We are currently running a major project funded by the EPSRC. This is exploring how energy harvesters and signal processing algorithms can be designed together. Our aim is to establish the trade-off between: physiological signal strength, motion artefact corruption of the collected signal, and energy harvesting potential; as we perform sensing on different parts of the body. 

Outputs

Keep up to date with our publications and high impact reseach outputs

Join us

Are you interested in working on next generation wearable devices? Our current activities are focusing on: energy harvester powered sensor nodes, new algorithm development, and the design of very low power fully custom microchips. If any of these areas interest you:

  1. Explore our funding opportunities
  2. Get in touch

Project main investigator: Dr Casson

▲ Up to the top