Chapter 2
Sampling and Reconstruction
2.1 Theory
WHENEVER we wish to obtain a real world signal in order to process it digitally, we must first convert it from its natural analog form to the more easily manipulated digital form.1 This involves grabbing, or “sampling,” the signal at certain instants in time. We assume the sampling instants are equally spaced in time (TS), so that the sampling frequency (Fs) is equal to 1/TS. Each individual sample represents the amplitude of the signal at that instant in time, and the number of bits per sample that we use to store this amplitude determines how accurately we can represent it. More bits means better fidelity, but it also means greater storage and processing requirements. The effect ...
