On-Board Feature Extraction from Acceleration Data for Activity Recognition

Atis Elsts, Ryan McConville, Xenofon Fafoutis, Niall Twomey, Robert Piechocki, Raul Santos-Rodriguez, Ian Craddock

Research output: Chapter in Book/Report/Conference proceedingConference Contribution (Conference Proceeding)

82 Downloads (Pure)


Modern wearable devices are equipped with increasingly powerful microcontrollers and therefore are increasingly capable of doing computationally heavy operations, such as feature extraction from sensor data. This paper quantifies the time and energy costs required for on-board computation of features on acceleration data, the reduction achieved in subsequent communication load compared with transmission of the raw data, and the impact on daily activity recognition in terms of classification accuracy. The results show that platforms based on modern 32-bit ARM Cortex-M microcontrollers significantly benefit from on-board extraction of time-domain features. On the other hand, efficiency gains from computation of frequency domain features at the moment largely remain out of their reach.
Original languageEnglish
Title of host publicationEWSN ’18
Subtitle of host publicationProceedings of the 2018 International Conference on Embedded Wireless Systems and Networks
PublisherAssociation for Computing Machinery (ACM)
Number of pages4
ISBN (Print)9780994988621
Publication statusPublished - 16 Feb 2018

Structured keywords

  • Digital Health


  • Digital Health


Dive into the research topics of 'On-Board Feature Extraction from Acceleration Data for Activity Recognition'. Together they form a unique fingerprint.

Cite this