GHLA

The GHLA algorithm builds on the activity classes found in the CREA algorithm, adding sensor calibration, and support for the aP4+ device.

Accelerometer calibration

The static acceleration sections in the datx file are used to calculate the calibration parameters for the acceleration sensor data to improve the accuracy of the inclination calculations.

Magnetometer calibration and heading calculations

The magnetometer data for the aP4+ device is calibrated to the unit sphere in order to calculate the compass heading during stepping.

Locus calculation

Loci are seperated using the following transition criteria (Fig 1):

  • Upright bout containing a continuous stepping bout >1 minute
  • Seated transport
  • Cycling

Fig. 1. Locus transitions: In the highlighted periods, the events meet the locus transition criteria, separating loci

The primary locus (Fig.2) is identified as all events between the last loci transition event of the previous day, and the first loci transition event of the next day. Secondary loci (Fig.3) are defined as between the other transition events in the day. Note: in this algorithm version returning to the primary locus during the day is not identified and is marked as a secondary locus.

Fig. 1. Primary loci: In the highlighted periods, the participant has not left their home

Fig. 3. Secondary loci: The highlighted periods are the remaining loci in between transition events. Note: currently returning to the primary locus during the day is not identified.

Algorithm versions

GHLA 2.2

Description of released algorithm as above

Official release (24th May 2022) in PALanalysis v8.11.8.74, PALbatch v8.11.1.62

Related Articles