(Dr. Wenjing Ye, advisor)

"Exploitation of Nonlinear Behavior to Improve the Performance of a Magnetic Sensor"

Abstract

This thesis is focused on the use of system non-linearities within a MEMS magnetic sensor to improve the sensitivity and therefore the performance of the sensor. An improved analytical model of a sensor design previously constructed within the E.C.E. is developed. The presentation of the model is followed with an optimization of the geometric parameters of the device that affect the sensor performance. The optimization of the passive system components is followed with the addition of actively powered components to improve sensor performance. The use of shaped comb drives is presented as means of increasing the non-linear behavior of the system by reducing the linear stiffness of the system and adding to the non-linear stiffness components. The addition of the shaped comb drives will improve the sensitivity of the sensor beyond what can be achieved through the optimization of the passive components. This is followed with the proposal of a sensing technique that uses the non-linear vibration behavior of the sensor as a means of characterizing the sensor and as a means of detecting changes in the magnetic environment around the sensor.