Translator Disclaimer
28 October 2006 A new time-frequency analysis method for gyro signal denoising
Author Affiliations +
Gyro output signal is a non-stationary and time-varied time series mixed with noise. A new time-frequency analysis method for gyro de-noising, the empirical mode decomposition (EMD), is introduced. The non-stationary hybrid gyro signal is studied adaptively in the domain of time and frequency. First the hybrid signal is decomposed into finite intrinsic mode functions (IMF), then their frequency characteristic and the relationship between these IMFs and different gyro errors are studied deeply. Follow that a certain number of IMFs are eliminated and EMD's filter bank characteristic is achieved. After the ring laser gyro signal is processed by EMD and wavelet, Allan variance approach is applied to evaluate five kinds of gyro noise. By the conclusion, the EMD method is obviously more efficacious than wavelet in eliminating various kind of noise.
© (2006) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Yanhong Lu and Yefeng Xu "A new time-frequency analysis method for gyro signal denoising", Proc. SPIE 6358, Sixth International Symposium on Instrumentation and Control Technology: Sensors, Automatic Measurement, Control, and Computer Simulation, 63580X (28 October 2006);


Back to Top