To effectively wipe out random drift and extract valid signal of MEMS gyro, the methods of adaptive Kalman filtering and wavelet analysis are investigated. For the first method, the autoregressive moving average (ARMA) model of random drift is established, which is essential to the adaptive Kalman filter. For the second one, the wavelet basis, decomposition level, and threshold-choosing principle are determined. Then the de-noising test is implemented by using real signal of MEMS gyro, and both methods are of good effectiveness. The contrast analysis between both methods indicates that the adaptive Kalman filtering approach is more suitable for the real-time de-noising of MEMS gyro signal.
Research on signal de-noising technique for MEMS gyro
01.06.2010
887109 byte
Aufsatz (Konferenz)
Elektronische Ressource
Englisch
Magnetically Suspended MEMS Spinning Wheel Gyro
Online Contents | 2005
|Magnetically suspended MEMS spinning wheel gyro
IEEE | 2005
|The gyro: a quartz MEMS automotive gyroscope
Kraftfahrwesen | 2006
|Real-time analysis for Stochastic errors of MEMS gyro
SPIE | 2017
|Praezise Richtungserkennung ohne Genauigkeitseinbussen - Gyro Sensoren auf MEMS Basis
Kraftfahrwesen | 2012
|