The Design of Low-Complexity Wavelet-Based Audio Filter Banks Suitable for Embedded Platforms

Smyth N. The Design of Low-Complexity Wavelet-Based Audio Filter Banks Suitable for Embedded Platforms. In: AES Convention 129. Audio Engineering Society; 2010. Paper 8260. Available from: https://aes.org/publications/elibrary-page/?id=15682

@inproceedings{Smyth2010_15682,
  author = {Smyth, Neil},
  title = {{The Design of Low-Complexity Wavelet-Based Audio Filter Banks Suitable for Embedded Platforms}},
  booktitle = {AES Convention 129},
  note = {Paper 8260},
  year = {2010},
  month = nov,
  publisher = {Audio Engineering Society},
  url = {https://aes.org/publications/elibrary-page/?id=15682}
}

TY  - CPAPER
TI  - The Design of Low-Complexity Wavelet-Based Audio Filter Banks Suitable for Embedded Platforms
AU  - Smyth, Neil
T2  - AES Convention 129
M1  - Paper 8260
PY  - 2010
DA  - 2010/11/06
UR  - https://aes.org/publications/elibrary-page/?id=15682
PB  - Audio Engineering Society
LA  - en
AB  - Many audio applications require the use of low complexity, low power and low latency filter banks (e.g. real-time audio streaming to mobile devices). The underlying mathematics of wavelet transforms provides these attractive characteristics for embedded platforms. However, commonly used wavelets (Haar, Daubechies) possess coefficients containing irrational numbers that lead to distortion in fixed-point implementations. This paper discusses the development and provides practical performance comparisons of filter banks using wavelet transforms as an alternative to more commonly used sub-banding filter banks in PCM audio coding algorithms. The advantages and disadvantages of wavelets used in such audio compression applications are also discussed.
ER  -