Least Squares Theory and Design of Optimal Noise Shaping Filters

Verhelst W, De Koning D. Least Squares Theory and Design of Optimal Noise Shaping Filters. In: AES Conference: 22nd International Conference: Virtual, Synthetic, and Entertainment Audio. Audio Engineering Society; 2002. Paper 000209. Available from: https://aes.org/publications/elibrary-page/?id=11160

@inproceedings{Verhelst2002_11160,
  author = {Verhelst, Werner and De Koning, Dreten},
  title = {{Least Squares Theory and Design of Optimal Noise Shaping Filters}},
  booktitle = {AES Conference: 22nd International Conference: Virtual, Synthetic, and Entertainment Audio},
  note = {Paper 000209},
  year = {2002},
  month = jun,
  publisher = {Audio Engineering Society},
  url = {https://aes.org/publications/elibrary-page/?id=11160}
}

TY  - CPAPER
TI  - Least Squares Theory and Design of Optimal Noise Shaping Filters
AU  - Verhelst, Werner
AU  - De Koning, Dreten
T2  - AES Conference: 22nd International Conference: Virtual, Synthetic, and Entertainment Audio
M1  - Paper 000209
PY  - 2002
DA  - 2002/06/06
UR  - https://aes.org/publications/elibrary-page/?id=11160
PB  - Audio Engineering Society
LA  - en
AB  - Signal requantization to reduce the word-length of an audio stream introduces distortions. Noise shaping can be applied to make requantization distortions minimally audible. The psychoacoustically optimal noise shaping curve depends on the time-varying characteristics of the input signal. Therefore, the noise shaping filter coefficients are to be computed and updated on a regular basis. We present a new theory for optimal noise shaping of audio. It provides more straightforward proof of properties of dithered and non-dithered noise shaping. In contrast with standard theory, it does show how optimum noise shaping filters can be designed in practice. Experimental results illustrate the method. The link with two popular noise shaping requantization techniques (F-weighted noise shaping and Super Bit Mapping) is also discussed.
ER  -