Signal processing is the process of turning the measured signals into useful information reflecting the essence of events of interest by extraction, transformation, synthesis, and display, such that engineers and scientists are able to analyze and optimize signals.
Transducers produce continuous analog time waveforms which have an infinite number of samples. Analog signals are commonly converted to digital signals via sampling, which are in the form of a discrete sequence of values. The digitized data is recorded in the time domain, but often it is desired to perform a Fourier transform to view and analyze the data in the frequency domain, i.e., displaying a parameter versus frequency. Since the real-world signals are almost inevitable to contain noise, averaging is usually performed in order to increase analysis accuracy, which can be done in both time and frequency domain.