On using coherence to measure distortion in hearing aids.

Area of Science:

• Signal processing
• Auditory system modeling

Background:

• Coherence is a frequency-domain measure assessing linear system input-output relationships.
• Signal-to-distortion ratio (SDR) can be calculated from coherence, but is prone to variance and bias.
• These inaccuracies affect the precise measurement of nonlinear effects and noise within a system.

Purpose of the Study:

• To investigate the origins of variance and bias in coherence estimates.
• To present novel procedures for mitigating these effects on distortion measurements.
• To demonstrate the effectiveness of these new methods using a simulated hearing aid.

Main Methods:

• Analysis of variance and bias sources in coherence estimation.
• Development of new signal processing techniques to reduce estimation errors.
• Validation of the improved methods on a simulated hearing aid response model.

Main Results:

• Identified key sources contributing to variance and bias in coherence-based SDR measurements.
• Demonstrated significant reduction in estimation errors using the proposed procedures.
• Validated improved accuracy in distortion measurements for simulated auditory systems.

Conclusions:

• The developed procedures effectively reduce variance and bias in coherence estimates.
• This enhances the accuracy of signal-to-distortion ratio (SDR) measurements.
• The findings have direct implications for evaluating nonlinear system performance, particularly in hearing aid technology.