Hi. The general theory states that "any periodic signal can be synthesized with a sum of harmonically related sinusoids". What if I have a signal f(x)=cos(2*pi*110x)+cos(2*pi*210x)? Clearly this signals contains the frequencies 110hz and 210hz, but they are not harmonically related? Is the fundamental frequency much much lower? What am I missing?
The arguments of the cosines and sines of a Fourier series are n (a x) where a is related to the periodicity of the function and n is the n
th harmonic. So, find an n
1 and n
2 such that
a = \(\displaystyle \dfrac{220\, \pi}{n_1}\, = \, \dfrac{420\, \pi}{n_2}\)
or
\(\displaystyle 220\, \pi n_2\, = \, 420\, \pi\, n_1\)
or
\(\displaystyle 11\, n_2\, = \, 21\, n_1\)
and we have n
1 is a multiple of 11 and n
2 is a multiple of 21 and we have
f(x) = \(\displaystyle cos(11\, (20 \pi x))+cos(21\, (20 \pi x))\)
Since a is sometimes written as \(\displaystyle \frac{2\pi}{L}\) where L is the length of the periodicity, you can see why Subhotosh Khan said "GCF of 110 and 210 is 10", i.e. if L is \(\displaystyle \frac{1}{10}\) then
f(x) = \(\displaystyle cos(11\, (\frac{2 \pi x}{L}))\, +\, cos(21\, (\frac{2 \pi x}{L}))\)
EDIT: Also note that these are the 11th and 21st harmonics for the f(x) given above. Had L been equal to 1, it would be the 110th and 210th harmonics. Thus the length of the periodicity is undetermined in the problem.