Stochastic resonance in MEMS capacitive sensors

Abstract

The paper presents a theory for stochastic resonance in MEMS capacitive chemical sensors and lays a framework for the calculation of sensor response. A parallel plate capacitive MEMS sensor biased to a fixed DC voltage and driven by a small amplitude AC voltage is considered in the presence of inherent cantilever noise sources. The stochastic resonance occurs due to correlation between the cantilever oscillations and the noise spectral components in the presence of dynamic nonlinearity. The noise power is transferred resonantly to the cantilever oscillations under certain conditions that are determined by the noise correlation rate, damping loss and AC drive frequency. The traditional strategies for enhancing sensor responses explore the optimization of cantilever geometry, DC bias level and AC drive amplitude and frequency. The present work provides theoretical framework for harnessing the noise power for enhancing the amplitude response of capacitive MEMS chemical sensors. The occurrence and usefulness of stochastic resonance in MEMS chemical sensor has been demonstrated by generating simulation results pertaining to explore the dependencies on various sensor parameters such as noise power and correlation rate, damping loss and polymer thickness.