In a log-log graph of diffused intensity I function of q (diffusion vector) you're supposed to get a linear function which slope equals the fractal dimension of the system you're studying. I get a linear slope but there are many big oscillations (goes up and down even though the overall behaviour is a slope that is going down) so knowing the slope is not easy. I also don't understand why it is not continuous. I am using the Fourier transform of the correlation function to get the structure factor (using the center of particles to get the correlation function) and to get the form factor using the points that describe the particle's geometry (and averaging over different sized to account for polydispersity, which should make the oscillations dissapear).
I also directly do the Fourier transform of the correlation function to get the diffused intensity direcly using all the points that describe my system (all the points, that describe many particles, in the simulation box).
