Nonlinear Behavioral Model of Capacitive MEMS Microphone for Predicting Ultrasound Intermodulation Distortion

Ultrasound Intermodulation Distortion (IMD) in microphones occurs when a system’s nonlinearity produces sum and difference products from a complex input signal containing two or more frequencies. In some cases, the distortion products can be audible even when the input signal contains only inaudible ultrasonic content, such as that from an ultrasonic ceiling occupancy detector. For this reason, it is beneficial for the hearing aid system to have minimum IMD. The main contributor to IMD is the MEMS transducer because of its nonlinear nature. This paper aims to study and investigate how design features affect intermodulation distortion using a nonlinear behavioral model/large signal model (LSM) of a capacitive MEMS microphone. The model-driven proposed MEMS microphone showed a 15 dB improvement of IMD comparing the base model at 20 kHz while retaining excellent overall acoustic performance.