光通信研究Issue(3)：28-36,9.DOI:10.13756/j.gtxyj.2026.260017
基于SCKF及子带处理的PCS系统偏振损伤均衡(特邀)
Polarization Impairments Equalization of PCS System based on SCKF and Subband Processing
摘要
Abstract
[Objective]To address the failure of traditional Multimode Algorithm(MMA)caused by rapid Rotation of State of Polarization(RSOP)and Polarization Mode Dispersion(PMD)in extreme thunderstorm scenarios,and to alleviate the high com-plexity of the current Kalman polarization impairment equalization scheme,this paper proposes a low-complexity two-stage polar-ization impairment equalization scheme based on Square-root Cubature Kalman Filtering(SCKF)and subband processing,which is designed for Probabilistic Constellation Shaping(PCS)Quadrature Amplitude Modulation(QAM)Polarization Division Multi-plexing(PDM)system considering practical application.[Methods]The proposed scheme innovatively integrates a low-com-plexity SCKF with subband processing mechanism.In the first stage,a large-step-size update strategy is adopted,incorporating the statistical characteristics of the PCS signal and an optimally designed threshold decision to construct a probability-aware inno-vation for a sliding-window SCKF.This achieves preliminary equalization of ultrafast RSOP and PMD.In the second stage,sub-band processing technology is introduced,where the full-band signal is firstly decomposed by an analysis filter bank into multiple subband signals.Then,the subband signals are processed in parallel using a low-tap Least Mean Square(LMS)algorithm to compensate for residual polarization impairments and Inter Symbol Interference(ISI).Finally,the signals are reconstructed through a synthesis filter bank to complete the polarization impairment equalization.[Results]Simulation results on a 32 GBaud PDM-PCS-64QAM system demonstrate that the proposed scheme effectively equalizes combined impairments of RSOP=18 Mrad/s with Differential Group Delay(DGD)=30 ps at 4.5 bits/symbol source entropy,and RSOP=13 Mrad/s with DGD=30 ps at 5 bits/symbol.The achieved performance is comparable to that of the conventional SCKF-based approach.At a higher entropy of 5.5 bits/symbol,it remains capable of handling RSOP=5 Mrad/s with DGD=30 ps,outperforming the bench-mark scheme.Benefiting from the large-step update strategy and the subband-parallel architecture,the computational complexity of the proposed method is reduced to only 62.19%of the traditional SCKF scheme,while lowering the required Optical Signal-to-Noise Ratio(OSNR)by more than 1.5 dB under various source entropy conditions.[Conclusion]In conclusion,the proposed scheme significantly reduces computational complexity while maintaining excellent joint polarization impairment equalization capa-bility and noise robustness,thereby providing a reliable solution for polarization impairment equalization in extreme scenarios.关键词
平方根容积卡尔曼滤波/概率星座整形/子带处理/最小均方算法Key words
SCKF/PCS/subband processing/LMS algorithm分类
信息技术与安全科学引用本文复制引用
乔京帅,许恒迎,白成林,许雅云,刘红,刘婷..基于SCKF及子带处理的PCS系统偏振损伤均衡(特邀)[J].光通信研究,2026,(3):28-36,9.基金项目
国家自然科学基金资助项目(62371216,62101229,61501213) （62371216,62101229,61501213）
山东省自然科学基金资助项目(ZR2025MS1018,ZR2022MF253,ZR2020MF012,ZR2020QF005) （ZR2025MS1018,ZR2022MF253,ZR2020MF012,ZR2020QF005）
