In-depth


Question 7.4

 

Quantization noise

Another very important parameter in any source encoding scheme is the level of noise and distortion introduced by the coding process. For waveform encoding, the main noise source is quantization error, that is, the amplitude errors which the A/D and D/A conversion process introduces into the signal by not having infinite precision. The level of quantization noise is dependent on how close any particular sample is to one of the M levels in the converter. For a speech input, this quantization error will be manifest as a noise-like disturbance at the output of a D/A converter.

The signal to quantization noise ratio for an A/D converter is thus clearly a function of the number of bits used and is given by (see in-depth):

Peak S/N ratio = 3M2 / 2

An 8-bit converter thus has a signal to quantization noise ratio of approximately 50 dB for a full-scale input signal.