Mitigating wind noise with a spherical microphone array.

Area of Science:

• Acoustics
• Signal Processing
• Noise Reduction

Background:

• Wind noise significantly corrupts sound measurements, especially in outdoor or airflow environments.
• Traditional noise reduction methods often degrade the quality of the desired sound signal.
• Spherical microphone arrays offer advanced spatial audio capture capabilities.

Purpose of the Study:

• To develop and validate a method for reducing wind noise using a rigid spherical microphone array.
• To differentiate and suppress wind noise from desired sound signals in the spherical harmonic domain.
• To assess the effectiveness of the proposed method in preserving sound pressure level (SPL).

Main Methods:

• Utilizing a rigid spherical microphone array for sound and wind noise measurement.
• Analyzing sound and wind noise in the spherical harmonic domain.
• Applying a low-pass filter in the spherical harmonic domain to reduce wind noise.
• Conducting experiments in an anechoic chamber with controlled sound and wind sources.

Main Results:

• Wind noise distribution in the spherical harmonic domain is irregular, unlike the concentrated sound signal.
• The proposed filtering method successfully reduced wind noise by over 10 dB below 500 Hz.
• Desired sound SPL was extracted with an error within 1.0 dB, even with 8 dB lower sound levels than wind noise.

Conclusions:

• The spherical harmonic domain filtering method effectively reduces wind noise while preserving the desired sound signal.
• This technique offers a robust solution for acoustic measurements in the presence of significant wind noise.
• The method demonstrates high accuracy in extracting target sound signals under challenging noisy conditions.