Research On Carrier Recovery Technology Of High Order Modulated Optical Signal Baesed On Kalman Filter

This paper mainly proposes a carrier recovery technique based on parallel operation linear Kalman filter(LKF)algorithm to estimate and compensate the corresponding frequency offset and carrier phase for various high-order modulation formats.The high-order modulation format involved.There are 16 QAM,64QAM and circular 16QAM;the algorithm is modified appropriately for each high-order modulation format to achieve the best carrier recovery effect;and the performance of the algorithm is evaluated by simulation.According to the 16 QAM modulated signal,a carrier recovery scheme based on extended Kalman filter(EKF)and parallel operation linear Kalman filter cascade is proposed.The EKF makes an initial estimation of the frequency offset of the signal,and can be completed with a small amount of signal data;then the residual frequency offset and phase noise are simultaneously estimated by using LKF,and the information is processed in the form of a data block,which greatly reduces the computational complexity of the algorithm.For the proposed algorithm,firstly analyze its optimal Q parameter selection and discuss the influence of different data block lengths on the algorithm;perform performance test on the proposed carrier recovery scheme,in which the frequency offset estimation can reach 12.5% of the baud rate;The simulation test of the tracking ability of the frequency offset drift is carried out,and the maximum tracking frequency drift speed reaches 320 MHz/us.For 64 QAM high-order modulation signals,this paper uses the Fast Fourier Transform(FFT)algorithm with better precision as the algorithm for initial estimation of frequency offset.The existing parallel block linear Kalman filter is preserved.The discussion and analysis of algorithm parameter optimization are carried out to get the best optimization parameters,so that the algorithm achieves the best results.Then,the simulation continues to test the frequency offset estimation performance and linewidth tolerance of the proposed algorithm under 64 QAM signal.The circular 16 QAM signals for two different distribution patterns are(4,4,4,4)and(1,5,5,5)respectively;the M-squares frequency offset estimation algorithm and the parallel operation of the block LKF Carrier recovery combines carrier recovery for the above two circular 16 QAMs.The scheme can ensure that the phase noise tolerance is also improved when the frequency offset estimation is high.In the nonlinear transmission scenario,the effectiveness of the above carrier recovery scheme is tested.The article uses VPI simulation software to build a dual-polarization fiber transmission link.It is concluded that the phase noise in the signal can be obtained by using the proposed scheme under nonlinear transmission.The compensation is given to give the best fiber-input power of the 16 QAM signal,and the farthest transmission distance can reach 1388 km.The nonlinear transmission performance under three different dispersion management links is discussed,and the comparison of the simulation results is obtained.The 16 QAM under dispersion management has an order of magnitude higher tolerance than the nonlinear tolerance under the dispersion management link.Therefore,the choice of DSP algorithm not only has a great advantage in cost,but also better performance.The carrier recovery algorithm proposed in this paper can well solve the effects of frequency offset and phase noise in many high-order modulation formats,and the algorithm has low computational complexity,fast computation speed,and simultaneous frequency offset and phase estimation and frequency offset estimation range.A wide range of advantages,such as a wide range of applications,with high practical applicability.