Content deleted Content added
m cat |
No edit summary |
||
Line 9:
where ''x'' is a real number, ''Q''(''x'') an integer, and ''f''(''x'') is an arbitrary real-valued function that controls the "quantization law" of the particular coder.
In computer audio, a linear scale is most common. here ''f''(''x'') = ''x-0.5''. The quantization operator can therefore be alternately expressed as,
For example, in digital [[telephone|telephony]], two popular quantization schemes are the '[[A-law algorithm|A-law]]' and '[[Mu-law algorithm|µ-law]]', each mapping an analog signal to an integer value represented by an 8-bit [[binary]] number, but each with a different function ''f''. ▼
:<math>Q(x) = \operatorname{floor}(x)</math>
where ''floor()'' returns the highest integer less than or equal to x. With this quantization law, the [[signal-noise ratio]] can be approximated as
:<math>\frac{S}{N_q} = (6.02M + 1.76)dB</math>
where M is the number of bits being used to code the audio. From this equation, it is often said that the SNR is approximately 6dB per bit.
▲For example, in digital [[telephone|telephony]], two popular quantization schemes are the '[[A-law algorithm|A-law]]' and '[[Mu-law algorithm|µ-law]]', each
See also:
| |||