A project I am engaged with involves taking raw accelerometer data (in g's ) and analyzing for the existence of tremors (the accelerometer is attached to an individuals hand). I am relatively new to digital processing, and was interested in the steps necessary to remove any noise and eventually convert acceleration to velocity and displacement measurements.

I imagine this would entail some kind of filter for the de-noising, but am pretty in the dark for the rest. Any insight would be appreciated! Thanks

  • $\begingroup$ It would be helpful with the dimensions of the data, like how many time series you have and the resolution. A simple plot of a piece of data is always nice too. Also check out digital signal processing they specialize on this type of questions! Good luck! $\endgroup$
    – ragulpr
    Commented Oct 17, 2016 at 20:58
  • $\begingroup$ Some advice on processing accelerometer data is provided here: stats.stackexchange.com/questions/40354/… $\endgroup$
    – BGreene
    Commented Oct 19, 2016 at 9:00


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.