Frequency-scanning interferometry for dynamic measurement using adaptive Sage-Husa Kalman filter

The measurement problem for dynamic targets under frequency-scanning interferometry (FSI) can be solved using Kalman filtering (KF). However, the FSI system is extremely sensitive to environmental changes, and the constant initial value of the KF filter can easily lead to filter divergence. In this paper, a novel adaptive Sage-Husa Kalman filter is proposed, which can solve the problem of filter divergence caused by the wrong selection of initial values and has strong robustness under the interference signal with low signal to noise ratio. It utilizes online stochastic model and improved Sage-Husa filter to update the noise variable. In addition, the principle of covariance matching is used to iterate over noise. The experimental results illustrate that the proposed algorithm has better robust and accurate results compared to existing filters.