A Robust Phase Unwrapping Algorithm With High Measurement Range Based on Unscented Kalman Filter in Interferometric Optical Fiber Sensing

Phase unwrapping is the prerequisite to obtain the true phase in interferometric fiber sensing. This article proposes a method based on an unscented Kalman filter (UKF) for interferometric fiber sensors. This method can break the limitation that the traditional phase unwrapping methods fail with a high probability when the differences in phase change are more than
$\pi $
radians according to the Itoh criterion. Then, the measurement range of the phase signal can be expanded. This method combines the true phase and the phase gradient into the state vector estimation for UKF. The feasibility of this method is verified theoretically and experimentally. In the specific experiment, the phase signal with a maximum absolute phase difference of
$1.5\pi $
(about 4.705) between adjacent sampling points can be recovered perfectly. The time and frequency domain results using this method show better performance. This method can effectively improve the detection range of the system and reduce the sampling rate to a certain extent, which is suitable for large-scale array systems.