Enhancing MEMS Microphone SNR with Dual-Polysilicon Differential Design

In the pursuit of high-fidelity voice and acoustic sensing, MEMS microphone designers continually push the limits of sensitivity, bandwidth, and especially signal-to-noise ratio (SNR). One of the most effective techniques to boost SNR while preserving a compact footprint is through differential capacitive design, particularly using dual polysilicon microfabrication platforms.

A recent study showcases the design and implementation of a differential MEMS microphone built on a trench-refilled MOSBE process, leveraging just two polysilicon structure layers and achieving impressive acoustic performance metrics.

What Makes Differential MEMS Microphones Superior?

Unlike traditional single-ended designs, differential MEMS microphones use paired electrodes to cancel out common-mode noise and environmental interference. This results in:

Improved sensitivity and SNR
Higher resistance to mechanical or electrical noise
Enhanced clarity for voice recognition systems and audio capture

The reported design cleverly integrates top and bottom diaphragms with corresponding backplates, forming two distinct capacitive sensing electrode pairs. Electrical separation is achieved using a Si₃N₄ isolation layer, allowing true differential readout with minimal added complexity.

Performance Highlights

📌 Fabricated microphone key specs:

Differential sensitivity: −40.5 dB (re: 1 V/1 Pa)
Single-ended sensitivity: ~−46 dB
Signal-to-noise ratio (SNR): Over 57.8 dB
Bandwidth: 50 Hz – 22 kHz (±3 dB)
Footprint: 800 µm diameter

These characteristics make the microphone highly suitable for:

Voice-activated assistants
Smart speakers and earbuds
Studio-quality portable audio devices
Hearing aids and assistive listening systems

🎧 Want to explore MEMS microphones optimized for high SNR and compact form factors?
Browse our MEMS microphone product catalog

Manufacturing Approach: Two Polysilicon Layers, One Smart Design

This design is fabricated using a trench-refilled MOSBE process which includes:

Two polysilicon structural layers
Two sacrificial SiO₂ layers
One electrical isolation layer (Si₃N₄)

Innovative mechanical features like U-shaped springs and central posts are included to ensure diaphragm stability, reduce structural stress, and maintain uniform vibration — all contributing to superior SNR performance.

Why This Matters

MEMS microphones are increasingly deployed in AI-enabled, battery-powered systems where noise immunity, clarity, and power efficiency are paramount. A differential architecture offers a scalable path to improve performance without dramatically increasing cost or complexity.

Related Research & External References:

Conclusion

This differential MEMS microphone design, built using a two-layer polysilicon process, delivers a powerful combination of high sensitivity, wide bandwidth, and excellent noise immunity. These improvements enable manufacturers to meet the growing demand for low-power, high-fidelity MEMS audio solutions in today’s smart and mobile electronics landscape.

💡 Learn more about SISTC’s innovations in MEMS microphone technology:
👉 https://sistc.com/product-category/mems-microphone/