Page 124 - MODELING OF ASPHALT CONCRETE

102 Cha pte r F o u r

FIGURE 4-10 Examples of different imperfections of the sinusoidal test data. (Pellinen and Crockford
2003, with permission from RILEM Publications.)

Fast Fourier Transform
The following short introduction of signal processing and Fast Fourier Transform (FFT)
has been summarized from the Fundamentals of Signal Analysis by Hewlett Packard
(1989) and from a paper by Ramsey (1975).
The FFT is an algorithm for transforming data from time domain to frequency
domain. A finite number of data points need to be sampled using discrete intervals of
time separated by Δt. Then, time record is N equally spaced samples of input signal and
m
the requirement for the Fast Fourier analysis is that the dataset have exactly N = 2 data