Genetic Algorithm for Electro-Mechanical Co-Optimization of a MEMS Accelerometer Comprising a Mechanical Motion Pre-Amplifier with a 2Nd-Order Sigma Delta Modulator

Chen Wang,Huafeng Liu,Yuan Wang,Xiaoxiao Song,Jian Bai,Michael Kraft
DOI: https://doi.org/10.1109/dtip.2019.8752896
2019-01-01
Abstract:This paper describes a novel, semi-automated methodology based on a genetic algorithm (GA) using mechanical and electronic co-optimization for micro-electromechanical systems (MEMS) devices. The use of co-optimization facilitates a MEMS design flow at system level in which mechanical and electronic domains interact to achieve better system performance. A MEMS accelerometer comprising a mechanical motion amplifier with a 2nd-order electromechanical sigma-delta modulator (EM-ΣΔM) interface is presented to demonstrate the effectiveness of the design approach. The strength of the approach is verified by a simulated SNR performance of 69.8 dB. Preliminary open loop measurement results match well with simulation results in which the product of sensitivity and bandwidth is improved by 100% and the sensitivity by 141% compared to a device designed in a conventional way.
What problem does this paper attempt to address?