MEMS (microelectromechanical systems) microphones are micro-scale devices that provide high fidelity acoustic sensing and are small enough to be included in a tightly-integrated electronic product. It is no surprise that they can be found in smartphones and other consumer products such as smart speakers and headsets. Nowadays, MEMS microphones are not only used to record plain ambient sound, but they support stereo capabilities, active noise cancellation, directivity (through beam forming), voice recognition and other capabilities. These audio features are implemented by multiplying the number of microphones per device. For example, the latest smartphones include up to 6 MEMS-based microphones. The large range of uses for MEMS microphones has created substantial demand for these high-performance devices.
All microphones (conventional and MEMS-based) sense acoustic waves using a flexible membrane. The membrane moves under pressure induced by acoustic waves. Today, most MEMS microphones on the market use capacitive technology* to measure sound. Capacitive MEMS microphones work by measuring the capacitance between a flexible micro-scale membrane and a fixed backplate. Changes in air pressure created by sound waves cause the membrane to move. The backplate is perforated to let the air flow through it and is designed to remain rigid since the air can pass through its perforations. As the membrane moves, the capacitance between the moving membrane and the fixed backplate change (since the distance between them changes), and this change in electrical response can be analyzed and recorded.
There are several variations of MEMS-based capacitive microphones, such as
A MEMS microphone designer will want to study and optimize key performance metrics such as frequency response, sensitivity, the signal-to-noise-ratio (SNR), total harmonic distortion and input equivalent noise. The signal-to-noise ratio is a key performance measure, and different variations of MEMS capacitive microphones aim to increase the SNR by increasing the signal (using a double backplate and double membrane) or reducing the noise (using a sealed vacuum between two membranes).
Capacitive MEMS microphones and their performance characteristics can be designed, modeled and studied using MEMS+®, a component of the CoventorMP® MEMS design platform. MEMS+ supports the design of MEMS microphones by providing parametric, non-linear and multi-physics models of individual MEMS structures that can be assembled into a completed MEMS microphone design. Moreover, the integration of a MEMS+ microphone design into a Cadence Virtuoso® circuit offers the unique possibility to simulate the MEMS Microphone and its ASIC using specific IC biasing conditions. Please see our prior blog to learn more about using MEMS+ in microphone design.
Today, in the era of artificial intelligence, we are seeing new MEMS design strategies that use innovative automated optimization techniques. For example, a group at the Institute for Electronic Design Automation at the Technical University of Munich has studied and demonstrated the automated optimization of a MEMS microphone design in MEMS+ including its readout circuit1. We hope to discuss this fascinating topic at further length in a future blog post.
If you’re interested in including a MEMS microphone in your next product, please visit our MEMS Microphone Solutions page or Contact Us to learn more about how our tools can help you optimize your microphone design.
* Note that some MEMS microphones use piezoelectric technology, where a thin piezoelectric layer and its electrodes are directly bonded to the top of the membrane. When the piezoelectric layer is moved by the pressure of sound waves, an electrical signal is produced due to the piezoelectric effect. This configuration can also be studied with CoventorMP.