Equalizing the In-Ear Acoustic Response of Piezoelectric MEMS Loudspeakers Through Inverse Transducer Modeling

Area of Science:

• Acoustics
• Electrical Engineering
• Signal Processing

Background:

• Micro-Electro-Mechanical Systems (MEMS) loudspeakers are gaining traction for in-ear audio due to their miniature size.
• Pronounced resonances in MEMS loudspeaker frequency response limit their practical application.
• Conventional miniature transducers face limitations in modern audio devices.

Purpose of the Study:

• To develop a model-based digital signal equalization approach for MEMS loudspeakers.
• To mitigate the impact of mechanical and acoustic resonances on MEMS loudspeaker performance.
• To achieve a flat Sound Pressure Level (SPL) frequency response for improved audio fidelity.

Main Methods:

• A circuit equivalent model of the MEMS loudspeaker was utilized.
• An inverse circuital model was constructed using nullors.
• Discrete-time implementation was achieved using Wave Digital Filters (WDFs).
• The inverse system pre-processed the input voltage signal to compensate for transducer characteristics.

Main Results:

• The proposed digital equalization method significantly flattened the Sound Pressure Level (SPL) response.
• Compensation was effective across the 100 Hz-10 kHz frequency range.
• The maximum deviation from a target flat frequency response was less than 5 dB.

Conclusions:

• Model-based digital signal equalization effectively addresses resonance issues in MEMS loudspeakers.
• The Wave Digital Filter implementation provides a practical solution for real-time audio signal processing.
• This approach enhances the suitability of MEMS loudspeakers for high-fidelity miniature audio applications.