Improving the Performance of MEMS-based Inertial Sensors by Removing Short-Term Errors Utilizing Wavelet Multi-Resolution Analysis

W. Abdel-Hamid, A. Osman, A. Noureldin, and N. El-Sheimy

Abstract:	Most of the present INS/GPS integration techniques utilize Kalman filtering which suppresses inertial sensors noise but within a limited band at high frequency. If the same technique is applied for MEMS-based INS, the overall accuracy may be jeopardized due to the significance of its inertial sensor errors that are mixed with the motion dynamics. This research work aims at improving the performance of MEMS-based inertial sensors utilizing wavelet multi-resolution analysis (WMRA). WMRA has been implemented to improve the sensors signal to noise ratio, remove sensor errors mixed with motion dynamics and provides more reliable data to Kalman-filtering based INS/GPS integration method. This can be achieved in two steps. The first step involves eliminating the high frequency sensor noise using wavelet de-nosing methods. The second step is thresholding which specifies a proper threshold level at which the motion dynamics can be separated from the short-term and/or long-term sensor errors as well as other disturbances like engine vibrations, pavement and tire irregularities. The second step involves decomposing the inertial sensor signals at different levels of resolution while applying appropriate thresholding criterion. Effectiveness and validity of the proposed method were investigated through a kinematic van test using integrated GPS and MEMS-based inertial measurement unit (IMU). The integration of MEMS-based IMU and DGPS measurements was performed using conventional Kalman filter, before and after wavelet analysis. Various intentionally introduced GPS outage periods were considered during data processing. The position components are compared to a reference trajectory, provided by DGPS, to examine the performance of the MEMS inertial sensors. The results showed the effectiveness of the WMRA technique in improving the performance of MEMS-based inertial sensors.
Published in:	Proceedings of the 2004 National Technical Meeting of The Institute of Navigation January 26 - 28, 2004 The Catamaran Resort Hotel San Diego, CA
Pages:	260 - 266
Cite this article:	Abdel-Hamid, W., Osman, A., Noureldin, A., El-Sheimy, N., "Improving the Performance of MEMS-based Inertial Sensors by Removing Short-Term Errors Utilizing Wavelet Multi-Resolution Analysis," Proceedings of the 2004 National Technical Meeting of The Institute of Navigation, San Diego, CA, January 2004, pp. 260-266.
Full Paper:	ION Members/Non-Members: 1 Download Credit Sign In