Research On The Rotation Of State Of Polarization Tracking And Compensation Algorithms In Coherent Optical Communication Systems

The increasing advancement of mobile communication,the Internet,and cloud computing technologies has fueled the need for high-speed,high-capacity,and high-reliability communication.In this regard,coherent optical communication technology has become a popular choice due to its high transmission rate,strong anti-interference capabilities,and high security.However,the polarization multiplexed optical communication system is now facing irreparable impairment due to the rotation of the state of polarization(RSOP)in the actual optical fiber link.The conventional constant modulus algorithm has proven to be insufficient in tracking highspeed RSOP.To address this issue,the Kalman filter offers two benefits of fast convergence and high dynamic tracking performance.It is a commonly used technique in optical fiber communication systems for RSOP tracking and correction.Therefore,there is significant theoretical significance and real-world application value in the research on RSOP tracking and compensation algorithms in coherent optical communication systems.This thesis investigates the rapid RSOP effect in fiber-optic channel transmission,including the research on the adaptive Extended Kalman Filter algorithm based on a residual decision,the adaptive Cubature Kalman Filter algorithm based on average decision error covariance,and the adaptive strong tracking Cubature Kalman Filter algorithm.The main research work is summarized as follows:(1)This thesis proposes an adaptive Extended Kalman Filter algorithm based on a residual decision to address the issues of frequent tuning parameter changes,high computational cost,and overall poor system performance in the adaptive Extended Kalman Filter.The proposed algorithm employs diagonalized covariance matrices and residual decision detector to prevent frequent tuning parameter updates.Simulation results demonstrate that compared with the diagonal Extended Kalman Filter algorithm,the proposed algorithm can achieve two orders of magnitude reduction in BER performance by adaptively updating the tuning parameters while satisfying the forward error correction(FEC)threshold of 7%overhead is met.Moreover,under the same circumstances,it can withstand greater RSOP speeds with a maximum tracking azimuth rate of 130 Mrad/s compared to the adaptive Extended Kalman Filter algorithm.(2)This thesis proposes an adaptive Cubature Kalman Filter algorithm based on the average decision error covariance to address the problem of subpar adaptive performance of tuning parameters in the Cubature Kalman Filter technique.To dynamically update the tuning parameters,the system determines the average decision error covariance using the recovered signal.Simulation results show that the proposed algorithm can adaptively adjust the tuning parameters in various scenarios,which improves the adaptive performance of the algorithm.Compared to the Cubature Kalman Filter algorithm,the BER performance is reduced by two orders of magnitude under the condition of FEC threshold of 7%overhead.(3)This thesis proposes an adaptive strong tracking Cubature Kalman Filter algorithm to address the ultra-high-speed abrupt RSOP disruption brought on by various natural or artificial sources in real-world optical communication applications.The proposed algorithm dynamically updates the error covariance matrix using a fading factor to adjust the filtering gain.Simulation results demonstrate that the proposed algorithm can still meet the FEC threshold of 7%overhead and remain stable even when tracking to azimuth change rate of over 100Mrad/s.