Acceleration to Amplitude versus time

Hello,

I have accelerometer data of acceleration versus time, and I need to integrate it twice to get amplitude versus time to eventually use the fourier transform to get a frequency. Currently, when I use cumtrapz to integrate the data it will always give me a pretty straight line. I need it to be something that looks like a sinusoid. Can you explain why this is happening/what I can do to fix it?

Also, how would I do the fourier transform on something that is an array with time and amplitude components (it seems as though fourier only takes in a single vector)?

Answer 1

[Amal George M]

It looks like the acceleration value is always positive. If appropriate, consider negating the effect of gravity from the data.