A Research On Fiber Channel Estimation Division Based On Kalman Filtering

The progress of technology and the development of society have brought great changes to people's life.While advanced communication brings convenience to people,it also increases the demand for communication bandwidth and transmission rate,posing severe challenges to researchers of the field of optical communication.Coherent optical communication technology has attracted attention of researchers at home and abroad due to its advantages such as larger bandwidth,higher spectral efficiency,larger transmission distance and higher reception sensitivity;as well as the increasing possibility of its commercialization.Furthermore,Kalman filtering is applied to coherent optical communication system due to its excellent filtering performance.Kalman Filtering(KF)has such advantages as fast convergence speed,high estimation accuracy,wide application range and low complexity.Combining KF and coherent optical communication technology is an important research direction in the field of optical communication and an important means to improve the performance of the DSP algorithm of coherent optical receiver.Based on coherent optical communication system and KF,this thesis analyzes the basic principle of KF combining with the model of coherent transmission system,and applies KF to the compensation algorithm of partial impairment effect in transmission channel.For OFDM simulation system,Extend Kalman Filtering(EKF)is used to monitor and compensate the optical fiber dispersion of the system,and the influence of process noise covariance Q and measurement noise covariance R on KF estimation performance is also explored.On the basis of the three KF-based Carrier Phase Estimation(CPE)methods verified in the QPSK and 16 QAM simulation systems,the Adaptive Cascaded Covariance Matching(ACCM)algorithm ACCM-CPE is obtained by making improvements,which greatly improves the tolerance to the Q and R and is verified in the 10GS/s QPSK experimental system.Finally,the joint estimation algorithm for phase noise and amplitude noise CPANE and the improved algorithm ACCM-CPANE are analyzed and verified in the 10GS/s QPSK experimental system.The ACCM-CPANE greatly improved the tolerance of Q and R and improved the estimation performance.The improved polarization demultiplexing algorithm MKF based on polarization demultiplexing algorithm using traditional KF CKF is studied for polarization multiplexing PDM-QPSK and PDM-16 QAM simulation systems,and the feasibility of MKF is analyzed and verified in the single-carrier simulation system and OFDM experiment with faster convergence speed and better tolerance for Q and R than CKF.Secondly,for joint phase noise estimation and polarization tracking algorithm ECKF,three improved joint phase noise estimation and polarization demultiplexing algorithms EMKF,LCKF and LMKF are obtained based on the improvement of EKF and ECKF.The feasibility of LMKF,LCKF and EMKF are verified theoretically and analyzed by PDM-16 QAM simulation system with faster Rotation of States of Polarization(RSOP)tracking rate and better recovery performance for polarization state than ECKF.