Measure a force plate signal when a healthy adult is standing still on the plate, this is your signal. Measure an empty signal with nothing on the plate, this is your noise. SNR can be calculated by singal/noise, then square it, which is power ratio of signal/noise. If you want SNR in dB, apply 10 x log_10 {(signal/noise)^2}, or 20 x log_10 (signal/noise).

If your forceplatform is calibrated you do not need to measure this item. It would make sense if you are making a force platform and you want to measure it's function.

Dear Naghmeh Gheidi, two main techniques are normally used to evaluate the SNR:

a)-AT REST (as also proposed by Nam H Kim, from the State University of Jersey) : you record an empty signal with zero weight on the platform and a first estimate of the power of the noise is given simply by the sd (Standard Deviation) of the empty signal; this SNR can be assumed as due to the experimental equipment

b)-DURING MOTION: make a good estimation of the frequency band occupied by your force signal (i.e., by means of a carefull visual inspection of the force signal recorded during the dynamicl trials: you take just those parts of the signal that evidently you judge uncorrupted by the environmental noise. Let's say, for example, that you find that in your experiments the frequency band of the force signal is limited to 10 Hz). Since than ownwards you assume that all frequency content that will be found over 10 Hz can be attributed to noise (the so-called "noise outside the frequency range of the signal"). Therefore, by means of spectral estimation of the noise outside the usefull signal band you can have an estimate of the Signal-to-Noise Ratio:

SNR = 10*log_10(sd_signal/sd_noise)

This SNR can be assumed as due to environmental noise during motion experiment

Dear Filligoi,

Thank you for your complete and useful answer.

Best Wishes