A bulk acoustic mode single-crystal silicon microresonator with a high-quality factor

This paper details a bulk acoustic mode resonator fabricated in single-crystal silicon with a quality factor of 15 000 in air, and over a million below 10 mTorr at a resonant frequency of 2.18 MHz. The resonator is a square plate that is excited in the square-extensional mode and has been fabricated in a commercial foundry silicon-on-insulator (SOI) MEMS process through MEMSCAP. This paper also presents a simple method of extracting resonator parameters from raw measurements heavily buried in electrical feedthrough. Its accuracy has been demonstrated through a comparison between extracted motional resistance values measured at different voltage biases and those predicted from an analytical model. Finally, a method of substantially cancelling electrical feedthrough through system-level electronic implementation is also introduced. © 2008 IOP Publishing Ltd.

Direct parameter extraction in capacitively transduced micromechanical resonators using the anti-resonance

This paper presents a method for the fast and direct extraction of model parameters for capacitive MEMS resonators from their measured transmission response such as quality factor, resonant frequency, and motional resistance. We show that these parameters may be extracted without having to first de-embed the resonator motional current from the feedthrough. The series and parallel resonances from the measured electrical transmission are used to determine the MEMS resonator circuit parameters. The theoretical basis for the method is elucidated by using both the Nyquist and susceptance frequency response plots, and applicable in the limit where CF > CmQ; commonly the case when characterizing MEMS resonators at RF. The method is then applied to the measured electrical transmission for capacitively transduced MEMS resonators, and compared against parameters obtained using a Lorentzian fit to the measured response. Close agreement between the two methods is reported herein. © 2010 IEEE.

An approximate likelihood method for estimating static parameters in multi-target tracking models

This paper considers a class of dynamic Spatial Point Processes (PP) that evolves over time in a Markovian fashion. This Markov in time PP is hidden and observed indirectly through another PP via thinning, displacement and noise. This statistical model is important for Multi object Tracking applications and we present an approximate likelihood based method for estimating the model parameters. The work is supported by an extensive numerical study.