An Improved Empirical Mode Data Denoising Algorithm for Atomic Clocks
Affected by the internal hardware facilities of the instrument and external disturbances,the clock group frequency difference data of the atomic clocks will have noise and coarse difference values,which seriously affect the stability and accuracy of the atomic clock signals.So an improved denoising algorithm for atomic clocks is proposed,that is the improved EMD-AKF algorithm.After the atomic clock data are removed from the coarse difference according to the 3σ criteria,the preprocessed data are analyzed by combining empirical mode decomposition and adaptive Kalman filtering.The dominant components in the IMF components after empirical mode decomposition are determined according to the Pearson's correlation coefficient and the autocorrelation coefficient.The noise-dominant IMF components are denoised by adaptive Kalman filtering,and the new atomic clock clock difference data are finally reconstructed.The stability assessment of the clock difference data before and after processing shows that the improved EMD-AKF algorithm improves the frequency stability of the atomic clock signal by one order of magnitude,and greatly reduces the influence of the noise of counters and other instruments on the frequency difference data of the atomic clock.
NETL
NSTL
万方数据
