Sparse sound field decomposition for super-resolution in recording and reproduction

Area of Science:

• Acoustics
• Signal Processing
• Computational Physics

Background:

• Current sound field decomposition methods (plane-wave, harmonic) suffer from spatial aliasing artifacts.
• These artifacts limit the accuracy of sound field recording and reproduction, especially at higher frequencies.
• Existing techniques struggle to accurately represent complex sound fields near sources.

Purpose of the Study:

• To propose a new sound field decomposition method using sparse decomposition.
• To improve the accuracy and resolution of sound field recording and reproduction.
• To mitigate spatial aliasing artifacts inherent in conventional methods.

Main Methods:

• Developed a generative model for sound fields, incorporating near-field and far-field components.
• Utilized sparse decomposition algorithms to separate near-field (spatially sparse) and far-field components.
• Applied the decomposed components for enhanced sound field interpolation and super-resolution.

Main Results:

• The proposed method accurately interpolates the sound field.
• Achieved super-resolution in sound field recording and reproduction.
• Demonstrated improved reproduction accuracy above the spatial Nyquist frequency compared to existing methods.

Conclusions:

• Sparse sound field decomposition offers superior accuracy and resolution.
• The proposed generative model effectively captures complex sound field characteristics.
• This method overcomes limitations of traditional plane-wave and harmonic decomposition techniques.