Electrical Stimulation Based Statistical Calibration Model For MEMS Accelerometer And Other Sensors
Description
Micro Electro Mechanical Systems (MEMS) based accelerometers are one of the most commonly used sensors out there. They are used in devices such as, airbags, smartphones, airplanes, and many more. Although they are very accurate, they degrade with time or get offset due to some damage. To fix this, they must be calibrated again using physical calibration technique, which is an expensive process to conduct. However, these sensors can also be calibrated infield by applying an on-chip electrical stimulus to the sensor. Electrical stimulus-based calibration could bring the cost of testing and calibration significantly down as compared to factory testing. In this thesis, simulations are presented to formulate a statistical prediction model based on an electrical stimulus. Results from two different approaches of electrical calibration have been discussed. A prediction model with a root mean square error of 1% has been presented in this work. Experiments were conducted on commercially available accelerometers to test the techniques used for simulations.
Date Created
The date the item was original created (prior to any relationship with the ASU Digital Repositories.)
2020
Agent
- Author (aut): Bassi, Ishaan
- Thesis advisor (ths): Ozev, Sule
- Committee member: Christen, Jennifer Blain
- Committee member: Vasileska, Dragica
- Publisher (pbl): Arizona State University