View Full Version : Summarizing Triaxial Accelerometer Data

07-24-2008, 08:14 PM
Dear Biomech-L Subscribers,

My research group has recently branched out from using low resolution (sampling at 1Hz) accelerometric activity monitors (e.g., Actigraph (uniaxial), RT3 (triaxial) and Actical (omnidirectional)) to using +/-6g triaxial accelerometers (sampling at 80Hz). Although we are excited about the analytical possibilities these dense data provide, we still have an interest in comparing between the low vs. high resolution devices. As a result, I am searching for information on alternative methods of summarizing this 80Hz triaxial accelerometer data. At present it seems the most common method in the physical activity (i.e., energy expenditure) measurement field is to employ some form of vector magnitude (i.e., VM=sqrt(x^2 + y^2 + z^2).

If anyone has suggestions of other summarization methods or perhaps where to find this information (i.e., references) it would be much appreciated. In addition to wanting to learn about methods that combine axes, I am also interested in methods that somehow summarize the data yet maintain the triaxial nature of the data.

Thanks in advance for your time.

Kind regards,

Dale Esliger

Research Fellow
School of Sport and Health Sciences
University of Exeter
St. Luke's Campus
Heavitree Road
Exeter, EX1 2LU
England, UK
Email: d.esliger@exeter.ac.uk
Tel: +44 (0) 1392 262815
Fax: +44 (0) 1392 264726

SHS web site: http://www.exeter.ac.uk/exsport
CHERC web site: http://www.exeter.ac.uk/cherc