- #1

- 33

- 0

That parameter is somehow "perturbed" and its instantaneous value is:

y(t)= 123 +

sin(t - 50°) * 9 +

sin(t * 3 + 10°) * 3 +

sin(t * 20 + 60°) * 4

Suppose that I don't know the above formula, but I can calculate y(t) for any t. Hence I decide to use the Fourier transform to infer all the y's components.

If I use a sufficient number of samples to calculate y(t) (I currently use 2

^{16}samples) for 0 ≤ t ≤ 2π, I obtain:

y(t)= 122.999955606783 +

sin(t - 49.9942506502916°) * 8.99994072213981 +

sin(t * 3 + 10.0193463603278°) * 2.9999458244411 +

sin(t * 20 + 60.11052441823°) * 4.00000160503959

which is a very good result, while if I try to calculate y(t) for arbitrary t, I obtain a totally meaningless result.

Please, could somebody help me?

Thank you