Analog Signals and Digital Signals

Digital signals convey discrete symbols that are usually interpreted as digits. For example, a voltage that signals the numbers 1 through N by shifting between N distinct levels is a digital signal, and so is a sinusoid that signals N digits by shifting between N distinct frequencies or amplitudes. (The latter would be analog as regards its waveform, but digital as regards its signaling strategy.)

Most digital signals are binary; that is, they signal the digits 0 and 1 by shifting between two distinct physical states (e.g., a high voltage and a low voltage). Each 0 or 1 is a bit (binary digit). Other numbers are communicated by transmitting "words," bundles of 0s and 1s, either bit by bit along a single channel or in parallel (as by N wires all signaling at the same time). Typical word lengths are: 2⁴ = 16, 2⁵ = 32, and 2⁶ = 64 bits.

Although the term digital emphasizes the use of a finite number of signal states to communicate digits, it is really the use of such signals to convey symbols that makes digital signals uniquely useful. Whether the two states of a binary signal represent 0 and 1, Yes and No, "first half of alphabet" and "second half of alphabet," or any other pair of meanings is entirely up to the human designer. In principle, it is possible to use analog signals the same way, but in practice it is quite awkward to do so.

Another feature of virtually all digital systems is that all signals in the system change state (low to high, high to low) at frequent, regularly spaced instants. These system-wide changes of state are governed by a central timing device, or system clock. The system clock in a modern digital device may change state millions or billions of times per second.