Note: Many topics at this site are reduced versions of the text in "The Encyclopedia of Networking and Telecommunications." Search results will not be as extensive as a search of the book's CD-ROM.

ADC, or digitizing, converts analog waveforms to digital representations that can be processed and stored in computers. The analog wave is "sampled," or read, hundreds or thousands of times per second to map out the wave digitally. Digital music requires extremely high sampling rates (44,100 samples/sec), while it is usually acceptable to sample voice at 11,000 samples/sec or higher. There is also a factor that determines the precision of the captured signal-the more bits used to record the value of the sampled signal, the higher its resolution and the better its sound when played back. However, the more bits used, the more disk space is required for storage or bandwidth for transmission. For example, one minute of sampling at 44.1 kHz using 16 bits per sample (the compact disc specification) requires 5.292MB of disk space.

The telephone companies convert analog voice calls to digital at their central offices (there is one in your neighborhood) for transmission across trunk lines to other central offices or to long-distance systems. Voice converted to digital requires a 64-Kbit/sec channel, which happens to be a worldwide standard called DS0 (digital signal, level zero) for transmitting voice calls.

Analog-to-digital converters are used in a variety of information-processing applications. Information collected from analog phenomena such as sound, light, temperature, and pressure can be digitized and made available for digital processing. A codec (coder/decoder) is the device that transforms the analog signals to digital signals. The process involves sampling, quantizing, and digitizing. The amplitude of a signal is measured at various intervals. The tighter these intervals, the more accurate the recording. Figure A-3 illustrates how a wave is sampled 16 times per second, with a sampling rate of 16Hz. While sampling at this rate is impractical for voice or music, it illustrates how each sample records a different amplitude value for the sound. Generally, a rate of 8,000 samples per second or higher using 8 bits per sample is adequate for voice-quality signals. Quantizing is the process of replacing the sampled value with the nearest value within the range of the device and the sampling rate. Digitizing completes the process.