An Edge-Computing Sensor Platform for ISO 2631-1 Whole-Body Vibration Exposure Metrics

Timely feedback on whole-body vibration (WBV) exposure requires standardised metrics during measurement, yet many workflows record raw acceleration for offline processing, delaying feedback and increasing the data burden for bandwidth-constrained deployment. This paper presents a sensing node computing ISO 2631-1/AS 2670.1 exposure metrics on-device and transmitting metric records, not waveforms. The node combines an LIS2DH microelectromechanical systems (MEMS) accelerometer with an RP2040 microcontroller to calculate weighted root-mean-square (RMS) acceleration, daily exposure A(8), vibration dose value, maximum transient vibration value and crest factor. Algorithm-level verification showed that the implemented W_k filter reproduced the tabulated ISO 2631-1 response within 1.1% across 0.5-80 Hz. The node was compared under laboratory conditions with a CEM DT-178A datalogger whose 20 Hz recordings were reprocessed through a method-matched causal pipeline. Across ten trials per configuration under vertical excitation in the 6.3 Hz one-third-octave band, mean Z-axis A(8) differences were -4.5% (wired) and -7.9% (wireless). The datalogger's 20 Hz sampling was included within the comparison chain, and the lower realised W_k gain was consistent with the direction and approximate magnitude of the observed offset. Metric-level telemetry reduced the sustained payload rate by up to four orders of magnitude at the summary cadence, supporting bandwidth-constrained uplinks.