OFC 2026 中归入 Transport（相干、海缆、长途、多波段、SDM、空芯光纤、光网络智能化）的论文 225 篇，按二级专题分组；每篇标注技术层（网络/光系统/算法/器件/芯片）和场景。类型：PDP=Postdeadline，高分=Top-Scored。 返回：2026年 OFC 论文专题洞察
关键结论
单模光纤带宽拓到 OESCL 42.4 THz、>450 Tb/s〔Th4B.5〕；实时 2.5 Pb/s 24 芯〔Th3A.2〕；空芯光纤 0.040 dB/km〔M2J.1〕、266 km 超长跨段跨洋 21.7 Tb/s〔Th4B.7〕、单芯 550.97 Tb/s〔Th1J.6〕；相干可插拔 400G/λ × 5682 km 海缆〔Th4C.6〕、800G 多厂商互通 1602 km〔Th4B.6〕；网络智能化出现垂直大模型 Optics GPT〔Th4C.1〕与 OSFP 内实时纵向功率监测〔Th4B.4〕。
| 二级专题 | 论文数 | 范围 |
|---|---|---|
| HCF | 30 | 空芯光纤：制备、损耗、熔接、传输与部署 |
| AI光网络 | 56 | 数字孪生、LLM/智能体、ML、纵向功率监测与故障诊断 |
| 光系统建模 | 12 | GN/EGN、非线性与 SRS 模型、QoT 建模 |
| 高波特率器件 | 12 | 高波特率调制器/探测器/驱动/光 DAC |
| 光放与多波段 | 30 | EDFA/BDFA/拉曼/参量放大、S/U/E/O 多波段 |
| SDM光纤 | 22 | 多芯/少模/OAM 光纤与 SDM 传输 |
| 相干DSP与编码 | 23 | 相干 DSP、MIMO、概率整形、FEC、相噪/定时 |
| 光网络架构与控制 | 31 | ROADM/OXC、规划、SDN 控制、可插拔组网 |
| 光纤与测试 | 9 | 光纤本体、连接、测量表征 |
二级专题 × 场景
| 二级专题 | DCI | 城域 | 长途 | 海缆 | 通用 |
|---|---|---|---|---|---|
| HCF | 1 | – | 2 | 4 | 23 |
| AI光网络 | – | 7 | 3 | 1 | 45 |
| 光系统建模 | – | 1 | – | – | 11 |
| 高波特率器件 | – | – | – | – | 12 |
| 光放与多波段 | – | 1 | 4 | 4 | 21 |
| SDM光纤 | – | – | 3 | 2 | 17 |
| 相干DSP与编码 | 1 | 2 | 5 | – | 15 |
| 光网络架构与控制 | 5 | 2 | 2 | 1 | 21 |
| 光纤与测试 | – | – | 1 | 2 | 6 |
HCF
30 篇 · 空芯光纤：制备、损耗、熔接、传输与部署
| 编号 | 类型 | 技术层 | 场景 | 题目 | 中文要点 |
|---|---|---|---|---|---|
| M1J.1 | 高分 | 器件 | 通用 | Inline Measurement and Automated Feedback Control of Hollow Core Fiber Microstructure | DNANF空芯光纤拉制中微结构在线无损测量+反馈控制，制备20 km结构均匀低损光纤 |
| M1J.2 | 口头 | 器件 | 通用 | Impact of Higher-Order Modes on Mode Field Diameter Measurement in Hollow-Core Fibres | 系统研究高阶模对空芯光纤S²法模场直径测量的影响（尾纤、对准、摆放、端面旋转） |
| M1J.3 | 特邀 | 器件 | 通用 | High-performance fusion splicing technique for anti-resonant hollow-core fiber | 特邀：反谐振空芯光纤熔接技术、挑战与解决方案 |
| M1J.4 | 口头 | 器件 | 通用 | Fast, Low-Loss, and Field-Deployable Splicing of Anti-Resonant Hollow-Core Fibers | 侧视成像自动对准的空芯光纤熔接，97 s内最大损耗0.05 dB，30次100%成功，可现场部署 |
| M1J.5 | 特邀 | 器件 | 通用 | Making the Hollow Core Fibers Compatible with Current Fiber Infrastructure | 特邀：使空芯光纤兼容现有光纤基础设施的挑战与方案 |
| M2C.3 | 口头 | 光系统 | 海缆 | Real-time Unrepeatered Transmission of 400G/800G/1.2T over 726.1km/611.9km/436.1km HCF only Using EDFA | 仅用高功率EDFA，空芯光纤无中继实时传输400G/800G/1.2T分别达726.1/611.9/436.1 km（纪录） |
| M2J.1 | 特邀 | 器件 | 通用 | Low Intermodal Interference and Low Loss Hollow Core Fibers | 优化四重间隙管偏转角的GTA-ST-HCF，损耗0.040±0.003 dB/km，模间干扰−63.6~−73.2 dB/km |
| M2J.2 | 口头 | 器件 | 通用 | Backscattering Characterization in Hollow-Core Fibers: Isolating Air-Molecule and Surface Scattering Origins via Direct and Coherent OTDR Techniques | 直接+相干OTDR联合分离反谐振空芯光纤中空气分子与表面散射来源 |
| M2J.3 | 口头 | 器件 | 通用 | First Demonstration of Distributed Sensing Capability of NANF Hollow-core Fiber | 首次演示NANF空芯光纤的OFDR分布式应变/温度/弯曲传感，空间分辨率4 cm |
| M2J.5 | 口头 | 器件 | 通用 | Characterization of Multi-Path Interference of 10-km Hollow Core Fiber using Swept Wavelength Interferometry | 扫频干涉测量10 km支撑管空芯光纤冲激响应，揭示高阶模与多径干扰 |
| M4K.1 | 口头 | 算法 | 通用 | Extending Longitudinal Performance Monitoring to Hollow Core Fiber Systems: Feasibility and Implementation | 首次实验验证空芯光纤跨段的纵向性能监测，利用嵌入实芯段的非线性 |
| Th1A.6 | 口头 | 光系统 | 通用 | Dynamic evaluation of 2.32 Tbit/s signal and 10 W power co-transmission over 3 km AR-HCF | 3 km反谐振空芯光纤中2.32 Tb/s信号与10 W光功率共传，功率-速率-距离积17.4 W·Tb/s·km（纪录） |
| Th1J.2 | 口头 | 网络 | 通用 | Fiber Mixture Optimization of Future Hybrid HCF-SMF Multi-Span Transmission Systems | 混合HCF-SMF多跨系统光纤配比优化，可达与全HCF系统同等性能 |
| Th1J.3 | 口头 | 光系统 | 长途 | Leveraging Digital Subcarrier Multiplexing for Long-Haul Transmission over HCFs in the Presence of IMI | 低损空芯光纤长途传输中数字子载波复用较单载波更耐模间干扰(IMI) |
| Th1J.4 | 高分 | 光系统 | 长途 | 222-km-long Hybrid Span Transmission Systems made of Support Tube Hollow Core Fiber and of Standard Single Mode Fiber using High Power Doped Fiber Amplifier | 222 km长支撑管HCF+SSMF混合跨段系统，34 dBm高功率放大，>1000 km每通道AIR>800G |
| Th1J.5 | 高分 | 光系统 | 通用 | Transmitter Power Optimization for Uniform Performance Multi-Span Hollow-Core Fiber Transmission | 三跨442.7 km空芯光纤实时满载C波段32×800G，发射功率控制克服气体吸收，Q裕量>1.5 dB |
| Th1J.6 | 口头 | 光系统 | 通用 | Beyond 550 Tb/s S+C+L-band Bidirectional Transmission over 10.9-km Anti-Resonant Hollow-Core Fiber | 10.9 km反谐振空芯光纤S+C+L同波长双向393×2通道，净550.97 Tb/s（单芯单模光纤容量纪录） |
| Th2A.16 | 海报 | 器件 | 通用 | Novel Hollow Core Nodeless Antiresonant Leakage Inhibited Fiber with Low Confinement Loss over a Large Bandwidth | 新型无节点反谐振抑制泄漏空芯光纤设计，宽带低限制损耗、高阶模消光比高 |
| Th2A.50 | 海报 | 算法 | 通用 | Gas Line Absorption Mitigation in Hollow-Core Fibre using Spectral Pre-Equalisation | 频谱预均衡数字补偿空芯光纤CO₂气体吸收线，Q代价降低5.5 dB，优于383抽头均衡器1.3 dB |
| Th3A.3 | 特邀 | 器件 | 海缆 | Advanced SMART Techniques for the Hollow-Core Fiber Submarine ISAC | 特邀：空芯光纤海缆通感一体(ISAC)的先进SMART技术 |
| Th4B.7 | PDP | 光系统 | 海缆 | Sparsely repeated 21.7 Tb/s Net-Rate Transoceanic Transmission with 266 km Ultra-Long Spans Enabled by Low IMI and Low loss Hollow Core Fiber | GTA-ST-HCF空芯光纤266 km超长跨段跨洋6660 km传输净21.7 Tb/s，中继器<30个 |
| Th4B.8 | PDP | 器件 | 通用 | First Dual-Band Hybrid Window Antiresonant HCF with 0.13 dB/km loss at 1 µm and 0.11 dB/km at 1.55 µm | 首个双窗口超低损DNANF空芯光纤：1550 nm 0.11 dB/km，1015 nm 0.13 dB/km（纪录） |
| Tu3E.1 | 特邀 | 光系统 | 海缆 | Toward Multi-Petabit Submarine Cables: Opportunities and Challenges of Multi-Band, Multi-Core Fibers and Hollow-Core Fibers | 特邀：迈向多Pb/s海缆——多波段、多芯与空芯光纤的机遇与挑战，供电与空间约束决定可扩展性 |
| Tu3E.3 | 高分 | 器件 | 通用 | Accelerating Hollow-Core Fiber Deployment: An Efficient One-Shot Field Measurement Technique to Characterize CO2 Absorption-induced OSNR Penalty | 建模100 km空芯光纤CO₂吸收致OSNR代价，一次现场测量即可重新标定80/130 km HCF |
| Tu3E.4 | 口头 | 器件 | 通用 | Diagnostic Limitations of Unidirectional and Bidirectional OTDR in Hollow-Core Fiber Splicing | 单向与双向OTDR在空芯光纤熔接诊断中的局限性及组合方法 |
| Tu3E.6 | 特邀 | 光系统 | 通用 | Real-Time High Speed Transmission Over Deployed AR-HCFs: Challenges, Enabling Technologies, and Future Perspectives | 特邀：已部署反谐振空芯光纤实时高速传输——部署、架构演进、新信道损伤与波段选择 |
| W1H.1 | 特邀 | 网络 | DCI | Emerging Network-Wide Use Cases for Hollow-Core Fibers [Invited] | 特邀：空芯光纤全网级新兴用例——时延约束数据中心整合与高功率放大，及部署挑战 |
| W1H.2 | 高分 | 网络 | 通用 | Optimal Placement of Hollow-Core Fiber Spans in Optical Transport Networks with CAPEX Constraints | CAPEX约束下光网状网中空芯光纤跨段与高功率放大器的最优放置 |
| W1H.3 | 口头 | 网络 | 通用 | Optimal Output Power for C-band Optical Amplifiers in Transparent WDM Networks Based on Hollow Core Fiber | 空芯光纤透明WDM网络中C波段放大器最优输出功率：在线EDFA>32 dBm并非最优 |
| W2A.50 | 海报 | 算法 | 通用 | DSP Optimization for CO2 Absorption Impact in Hollow Core Fiber Coherent Optical Transmission Systems | 比较空芯光纤窄带CO₂吸收的DSP对策：预加重与数字子载波复用可降低代价 |
AI光网络
56 篇 · 数字孪生、LLM/智能体、ML、纵向功率监测与故障诊断
| 编号 | 类型 | 技术层 | 场景 | 题目 | 中文要点 |
|---|---|---|---|---|---|
| M2E.3 | 口头 | 算法 | 通用 | Bridging Static Modeling and Real-Time Control: A Hybrid AI-Physics Approach for EDFA Gain Dynamics | AI-物理混合模型实现EDFA增益动态实时优化，改善加/掉波过程瞬态 |
| M3A.1 | 特邀 | 算法 | 通用 | Dynamic ML Models for Evolving Networks: What, Where and How? | 演进网络中的动态ML模型：基于数据流更新离线模型，用于故障后流量预测与威胁检测 |
| M3A.2 | 口头 | 算法 | 通用 | A Unified Siamese Learning Framework for Zero-Day Anomaly Detection and Classification in Optical Networks | 多相似度孪生(Siamese)网络统一零日异常检测与单样本分类，准确率>99%，免重训练 |
| M3A.3 | 口头 | 网络 | 通用 | First Demonstration of Autonomous Fault Diagnosis with AI-Embedded Optical Modules in Passive WDM Systems for Front-Haul Network | 首次演示AI内嵌光模块在无源WDM前传中的自主故障诊断，四类故障分类准确率92.4% |
| M3A.4 | 口头 | 网络 | 通用 | Proactive Optical Network Defense: Leveraging Large Language Models for Failure Prediction and Maintenance | 基于LLM的光网络多变量故障预测与主动防御建议生成，准确率97.61% |
| M3A.5 | 口头 | 算法 | 通用 | Multi-factor conditional DDPM for high-fidelity data augmentation in optical network soft failure diagnosis | 多因子条件扩散模型(DDPM)做数据增强，仅用10%训练数据达全量81%的软故障诊断性能 |
| M3A.6 | 口头 | 算法 | 通用 | Conditional Denoising Diffusion Probabilistic Model for Trustworthy Failure Prediction in Optical Networks | 条件扩散模型可信故障预测，F1达97.42%并给出置信度与不确定度量化 |
| M3A.7 | 口头 | 算法 | 通用 | Experimental Demonstration of Online Learning-Based Concept Drift Adaptation for Failure Detection in Optical Networks | 在线学习适应概念漂移的光网络故障检测，性能较静态模型提升最多70% |
| M3K.2 | 口头 | 算法 | 通用 | Dual-Timescale Meta-Learning for Few-Shot and Online EDFA Gain Prediction in Optical Network | 双时间尺度元学习EDFA增益预测，新器件仅需4个样本、误差0.19 dB，支持在线校准 |
| M3K.3 | 特邀 | 算法 | 通用 | Machine Learning Tools for Modeling and Optimization in Optical Communications and Optical Computing | 特邀：光通信与光计算建模优化的ML工具：黑盒、白盒与物理信息灰盒三种范式 |
| M3M.2 | 口头 | 算法 | 通用 | Experimental Semiconductor Optical Amplifier Dataset for Machine Learning | 两个SOA的增益/频谱/非线性实测数据集（多输入功率、电流、温度），附高斯过程回归示例 |
| M3Z.3 | Demo | 算法 | 城域 | Digital Twin–Driven Disaggregated Optical Network Reconfiguration and Optimization with Automated Conflict Resolution | 数字孪生驱动的解耦光网络重构优化与自动冲突消解（QoT优化+动态RWA） |
| M3Z.4 | Demo | 网络 | 城域 | Demonstration of an On-Prem Conversational AI Assistant for Unified Network Operations and Observability | 本地部署LLM多智能体对话助手统一IP/光网络运维与可观测性 |
| M3Z.5 | Demo | 网络 | 通用 | Scaling Optical Testbed Data Space for Data Sharing Across Vendor and Organizational Boundaries | 跨厂商/机构的光测试床数据空间，主权化数据共享支撑ML与预测性维护 |
| M3Z.6 | Demo | 网络 | 通用 | Intent-Driven Slicing Across Federated Compute and Optical Transport Networks | 意图驱动跨联邦算力与解耦光传送网切片，3GPP意图→QoT验证切片零接触部署 |
| M3Z.10 | Demo | 网络 | 城域 | Autonomous Capacity Scaling in Optical Metro Access Networks via Hierarchical SDN Control: A Demonstration | 分层SDN控制器利用跨域遥测自主触发城域接入光层扩容、化解拥塞 |
| M3Z.11 | Demo | 算法 | 城域 | Energy-Efficient Multiband Metro Network Emulator with Physical-Layer-Impairment-Aware Digital Twin | OpenROADM网络仿真器+物理损伤感知数字孪生，近实时预测拉曼倾斜，200通道C+L |
| M4A.1 | 特邀 | 算法 | 通用 | Practical Strategies Leveraging AI and Digital Twins for the Design and Management of Future Optical Networks | 特邀：结合AI与物理的数字孪生用于未来光网络设计与管理的实用策略 |
| M4A.3 | 口头 | 算法 | 通用 | Policy-driven Conformal Prediction for Trustworthy QoT Estimation | 策略驱动的保形预测(Conformal) QoT估计，统计保证可信，开放数据集准确率92%→99.6% |
| M4A.4 | 高分 | 网络 | 通用 | A Topology-Scalable Foundation Model for Network-Level QoT Estimation Using a Variable-Length Transformer | 可变长Transformer拓扑可扩展QoT基础模型，零样本MAE 0.68 dB，少样本微调误差降96% |
| M4A.5 | 高分 | 算法 | 通用 | OSNR/GSNR Prediction in Brownfield Links via a DLM-Anchored Hybrid Physics/ML Model | DLM锚定物理/ML混合模型预测存量链路逐通道功率/OSNR/GSNR，误差≤0.39/0.43 dB |
| M4A.6 | 口头 | 网络 | 通用 | LLM-Enhanced Digital Twin Framework in Optical Networks | LLM增强光网络数字孪生框架：意图解析+实时遥测+自动仿真+闭环优化 |
| M4K.3 | 口头 | 算法 | 通用 | PIDT: Physics-Informed Digital Twin for Optical Fiber Parameter Estimation | 物理信息数字孪生(PIDT)：参数化分步傅里叶+物理损失估计光纤参数，精度与收敛优于神经算子 |
| M4K.5 | 口头 | 算法 | 通用 | High-resolution Filtering Offset Detection and Localization using Longitudinal Power Monitoring | 相关型纵向功率监测定位偏心滤波器，参考波形扫描检测<3 GHz偏移 |
| Th1I.1 | 口头 | 算法 | 长途 | In-Service Laser Anomaly Detection via DSP-Enabled Optical Performance Monitoring | 基于DSP数据的在线激光器异常检测(OPM)，100 GBd 1600 km实验线宽估计误差<7% |
| Th1I.2 | 口头 | 算法 | 通用 | Continuous-Oriented Edge Flow Graph Model for Root Alarm Identification in Optical Network | 连续导向边流图模型建模告警因果传播方向，根告警识别准确率0.9795 |
| Th1I.3 | 口头 | 网络 | 通用 | First Field-Operational GraphRAG Agent for Information Query in Large-Scale Hierarchical Optical Networks | 首个现网运行的GraphRAG智能体，用于大规模分层光网络实时信息查询，自然语言→Gremlin准确率96.25% |
| Th1I.4 | 口头 | 算法 | 通用 | Cross-Operator and Security-Preserving Federated QoT Estimation over Heterogeneous SDM Optical Networks | 跨运营商保密联邦学习QoT估计，国际试验在异构SDM光网络上准确估计QoT |
| Th1I.5 | 口头 | 算法 | 通用 | Grey-box Digital Twin Model for QoT Estimation in Fielddeployed Networks | 中国电信现网灰盒数字孪生QoT模型（主干-分支网络），OSNR误差0.16 dB、功率误差0.294 dB |
| Th1I.6 | 口头 | 算法 | 通用 | AI-Assisted Identification and Characterization of Intra-Span Hybrid Fiber Types via Longitudinal Power Profile Estimation | 基于纵向功率分布估计的AI辅助识别跨段内混合光纤类型，零样本实验准确率90.9% |
| Th2A.19 | 海报 | 算法 | 通用 | Experimentally Characterized Digital Twin for Machine Learning Based EDFA Failure Prognostics | 实验表征数字孪生训练ML预测EDFA故障，准确率99.6% |
| Th2A.23 | 海报 | 算法 | 通用 | Transformer-Based Prognostics: Enhancing Network Availability by Improved Monitoring of Optical Fiber Amplifiers | 轻量Transformer从状态监测数据预测光纤放大器寿命，边缘预测性维护 |
| Th2A.25 | 海报 | 网络 | 通用 | UniOpt: A Unified Foundation-Model Approach for Scalable and Autonomous Management of Optical Networks | UniOpt：统一基础模型式光网络自主管理框架（多模态遥测+任务token化） |
| Th2A.33 | 海报 | 网络 | 通用 | On-Premises Small Language Model Agent with Physics-Aware Reasoning for Optical Network Optimization | 本地部署14B Qwen3小模型智能体+物理感知提示做光网络功率优化，接近GPT-4o |
| Th2A.52 | 海报 | 网络 | 通用 | A Scalable Approach to Longitudinal Power Profile for Embedded Systems | 面向嵌入式系统的内存高效纵向功率分布估计，>75%链路上精度代价<0.6 dB |
| Th2A.53 | 海报 | 算法 | 通用 | A Digital Twin Assisted Raman Amplifier Failure Detection and Re-Optimization in Multi-band Optical Networks | 数字孪生辅助正反向双网络检测多波段拉曼泵浦故障，2.5 s内在线重优化恢复GSNR |
| Th4B.4 | PDP | 算法 | 长途 | First Coherent DSP and Pluggable Transceiver Capable of Distance-Resolved, Longitudinal Power Monitoring | 首个在OSFP可插拔内实时相干DSP实现距离分辨纵向功率监测，800ZR+/400ZR+业务下1005 km范围、2.1 km分辨率 |
| Th4C.1 | PDP | 网络 | 通用 | Optics GPT: The First Vertically Pre-trained Foundation Model for Optics and Optical Communications | Optics GPT：首个光学与光通信垂直预训练基础模型(8B)，专业基准超GPT-4o与DeepSeek-R1，可生成DSP、估OSNR、自主修复故障 |
| Tu2A.3 | 海报 | 算法 | 通用 | Offset-Free Least-Squares Power Profile Estimation via DSP Frame Pilot Symbols | 利用DSP帧导频符号的无偏最小二乘功率分布估计，无需FEC后符号，80 GBd DP-16QAM验证 |
| Tu2A.5 | 海报 | 算法 | 通用 | Mode-Resolved Longitudinal Power and MDL Monitoring in Multi-Mode Fiber Transmission | 首个少模光纤中模式分辨纵向功率监测，定位分辨MDL低至1 dB |
| Tu2C.1 | 特邀 | 网络 | 通用 | Vision 2030: Autonomous Network Control in an All Optical World | 特邀：Vision 2030——全光世界中基于Agentic AI的自治网络控制（本地自治、智能体协同、灾难韧性） |
| Tu2C.2 | 口头 | 网络 | 城域 | Distributed and Dynamic AI Agent Collaboration over Optical Transport for Network Orchestration and Monitoring | 跨811 km光传送网分布式动态AI多智能体协同编排与监测，按实时条件选智能体，时延降2.9倍 |
| Tu2C.3 | 口头 | 算法 | 通用 | Machine Learning Assisted Digital Twin Framework for Improved EDFA Gain and QoT Estimation | GNPy+ML辅助EDFA模型的混合数字孪生，增益误差<0.1 dB，提升端到端SNR估计 |
| Tu2C.4 | 口头 | 算法 | 通用 | Experimental Validation of Lightpath Operation based on Continuous Digital-Twin Model Tuning | 多波段光网络中基于持续数字孪生模型调优的光路运维全流程，实验验证性能估计与劣化检测 |
| Tu2C.5 | 口头 | 算法 | 通用 | Differentiated Alarm Feature Modeling toward Intelligent Fault Recognition in Scale-Aware Optical Networks | 多算法协同的差异化告警特征建模实现光网络故障识别，准确率>98% |
| Tu2C.7 | 口头 | 算法 | 城域 | Reinforcement Learning Assisted Control Framework for Self-configuring SOA-Based Optical Metro-Access Networks | 强化学习(DDPG)闭环自配置SOA型无滤波器城域-接入网，提升OSNR并节能30% |
| Tu3A.5 | 口头 | 算法 | 通用 | Closed-form Expression for Spatial Correlation Function of Nonlinear Fiber Propagation and Its Applications to Longitudinal Power Monitoring | Nyquist信号非线性传播空间相关函数闭式表达，用于LPM：内存降99.999%并给出空间分辨率解析界 |
| Tu3E.2 | 口头 | 算法 | 海缆 | Extending a Single-Span Characterization to the Full Submarine Link: A Digital Twin Approach for Accurate G/OSNR Prediction over 150 spans | 数字孪生将单跨表征扩展到10200 km/150+放大器海缆，G/OSNR预测RMSE 0.23 dB |
| W1H.6 | 口头 | 算法 | 通用 | Adaptive-Transformer-Driven Transfer Learning for Online GSNR Optimization in Heterogeneous C+L-Band Networks | 带可训练适配器的预训练Transformer迁移学习，异构C+L网络在线GSNR优化，仅2%新数据误差0.34 dB |
| W2A.24 | 海报 | 算法 | 通用 | Network-Wide, Field-Deployable OSNR and Spectral Analytics Leveraging Transceivers as Diagnostic and Monitoring Tools | 商用收发机CW模式作零成本现场OSNR与频谱监测工具，~1 GHz带宽、OSNR精度≤1 dB |
| W2A.32 | 海报 | 网络 | 通用 | QoT-to-Configuration: Conditional VAE–cGAN for Few-Shot Inverse Design of Optical Networks | 条件VAE-cGAN少样本逆向设计：由QoT指标生成光网络配置 |
| W2A.49 | 海报 | 算法 | 长途 | Accuracy of LPM-based NLI Estimation over a Heterogeneous Link | 1483 km异构(SMF+PSCF)链路上基于LPM的NLI估计精度实验评估 |
| W2A.52 | 海报 | 算法 | 通用 | Optimal Spatial Step Size in Longitudinal Power Monitoring | 纵向功率监测最优空间步长选择，抑制衰减系数偏差并保持1 dB级损耗检测能力 |
| W3I.1 | 特邀 | 网络 | 通用 | Generative AI for Optical Networking: From Design Automation to Self-Operating Networks | 特邀（研讨会）：生成式AI用于光网络——从设计自动化到自运行网络（数字孪生、合成数据、LLM规划、意图到策略） |
| W4A.4 | 口头 | 网络 | 城域 | Forecast-Guided Reinforcement Learning for Extended-Access Metro P2MP Networks | 预测引导多智能体强化学习用于扩展接入城域P2MP网络动态资源分配 |
| W4I.1 | 特邀 | 网络 | 通用 | Agentic AI and Autonomous Networks: Shaping the Future of Connectivity | 特邀（研讨会）：自治网络与Agentic AI——意图驱动闭环自动化与差异化连接 |
光系统建模
12 篇 · GN/EGN、非线性与 SRS 模型、QoT 建模
| 编号 | 类型 | 技术层 | 场景 | 题目 | 中文要点 |
|---|---|---|---|---|---|
| M2C.2 | 口头 | 算法 | 通用 | Fast and Accurate ML-Based Nonlinear Interference Estimation Including SRS and PDL for Ultra-Wideband Transmissions | 毫秒级ML非线性干扰方差估计，含SRS与PDL，适用超宽带与色散管理配置 |
| M2C.4 | 口头 | 算法 | 通用 | Analytical QoT Model for Bi-Di Single-Fiber Point-to-Multipoint DSCM Coherent Systems | 单纤单激光器双向点到多点DSCM相干系统解析QoT模型，90%情况下精度0.3 dB |
| M3C.5 | 口头 | 算法 | 通用 | A Hybrid ISRS IS-EGN Model: Combining a Closed-Form Power Profile with Numerical Importance Sampling | 闭式功率分布+数值重要性采样的混合ISRS IS-EGN模型，高效评估C+L系统 |
| M3K.1 | 口头 | 算法 | 通用 | Physics-guided Scientific Discovery in EDFA Gain Modeling: A Closed-form Parameterization Approach | 物理引导科学发现得到闭式EDFA增益模型，外推MAE<0.35 dB，数据与复杂度低于神经网络 |
| M4K.7 | 特邀 | 算法 | 城域 | On the Potential of Optimized Perturbation Models for Fiber Nonlinearity Cancellation | 优化一阶正则微扰模型的光纤非线性消除潜力，单跨210–250 km PM-16QAM有效SNR增益>2 dB |
| Th2A.32 | 海报 | 算法 | 通用 | End-to-End GSNR Estimation While Securing Vendors’ Trade Secrets and Operators’ Confidentiality | 安全多方计算协议在不泄露各域参数下做解耦网络端到端GSNR估计，精度媲美GNPy |
| Th2A.54 | 海报 | 算法 | 通用 | Improved QoT Estimation through Input Parameter Refinement in Bidirectional Raman Amplified Links | 双向拉曼链路物理层模型输入参数精修，最大QoT估计误差降约2.91 dB |
| Tu3A.2 | 口头 | 算法 | 通用 | O-Band Closed-form Formula of the ISRS GN Model | O波段ISRS GN模型闭式公式（纳入FWM），O波段精度较前模型提升1.9 dB |
| Tu3A.3 | 口头 | 算法 | 通用 | Investigation of Wide-band Nonlinear Effects Leveraging a Semi-analytical EGN Model | 基于新功率演化表达的半解析EGN模型研究稀疏SRS均衡与符号率对宽带非线性的影响(C+L/S+C+L) |
| Tu3A.6 | 口头 | 算法 | 通用 | Fast and Stable Method for Simulation of Third-Order Backward Raman Amplifiers | 含ASE与瑞利散射的SRS边值问题快速稳定解法，仿真高功率三阶后向拉曼放大器并与实验比较 |
| W2A.51 | 海报 | 算法 | 通用 | Impact of Effective Area Tilt on Raman Gain Coefficient for Analytical SRS Modeling in UWB Transmissions | 考虑有效面积倾斜的频率相关拉曼增益解析模型，改进超宽带实时SRS功率演化建模 |
| W3J.3 | 口头 | 算法 | 通用 | O-Band Closed-form GN Model Verification with Wavelength-dependent System Parameters | O波段闭式GN模型（波长相关参数）实验验证，16.5 THz WDM 50.3/75.5 km，SNR预测误差<0.27 dB |
高波特率器件
12 篇 · 高波特率调制器/探测器/驱动/光 DAC
| 编号 | 类型 | 技术层 | 场景 | 题目 | 中文要点 |
|---|---|---|---|---|---|
| M2E.4 | 口头 | 器件 | 通用 | One-to-Ten Multicast of Self-injection Locking Laser with Sub-kHz Linewidth Enabled by Cross Phase Modulation | 交叉相位调制将100 Hz线宽自注入锁定激光器复制到10个75 GHz间隔波长（一分十组播） |
| M4D.2 | 口头 | 芯片 | 通用 | IQ Modulators with Two Segments using InP/SOI Chip-on-Wafer Bonding Process for Optical DAC | InP/SOI芯片-晶圆键合的两段式IQ调制器实现光DAC，带宽约60 GHz，64 GBd 16QAM |
| M4D.3 | 口头 | 芯片 | 通用 | High L-band Responsivity of Compact InP-on-Si Coherent-Receiver PICs via Chip-on-Wafer Bonding | 芯片-晶圆键合InP-on-Si紧凑相干接收PIC(1.2 mm²)，L波段高响应度，100 GBd DP-16QAM |
| Th1D.5 | 口头 | 芯片 | 通用 | Post Fabrication Permanent Tuning of Double Ring Filters for Narrow Linewidth Integrated Laser | 激光修调Si3N4双环滤波器有效折射率实现永久调谐，用于窄线宽集成激光器 |
| Th2A.43 | 海报 | 算法 | 通用 | Space-Polarization-Coded Coherent Receiver for LO Polarization-Insensitive Detection | 空间-偏振编码相干接收机，实现本振偏振不敏感检测，避免LO功率衰落 |
| Th3J.3 | 口头 | 算法 | 通用 | Practical Considerations for Laser Selection in DSP-free Coherent Systems Employing a Pilot-Tone-Assisted Optical Phase-Locked Loop | 导频辅助光锁相环免DSP相干系统中激光器选型框架，10 km空芯光纤130 GBd 16QAM验证 |
| Th3J.5 | 高分 | 器件 | 通用 | Silicon-Organic Hybrid (SOH) IQ Modulators with Sub-1 V π-Voltages Operating at 200 GBd 16QAM and 144 GBd 64QAM | Vπ<0.8 V硅-有机混合(SOH) IQ调制器，85 km传200 GBd 16QAM/144 GBd 64QAM（硅光IQ调制器纪录） |
| W1E.3 | 高分 | 芯片 | 通用 | Real-Time 1.2 Tb/s S-Band Silicon Photonics Transceiver Operating at 6-THz Tunable Bandwidth | S波段实时1.2 Tb/s硅光收发机，外腔ITLA 6 THz可调、>17.5 dBm，OSNR容限媲美C/L |
| W2A.2 | 海报 | 芯片 | 通用 | On-Chip Wavelength Monitoring in a Widely Tunable Hybrid InP/Polymer Photonic THz Source | 混合InP/聚合物可调THz光源片上波长监测（薄膜滤波器+PD），0.5 THz范围连续调谐 |
| W2A.11 | 海报 | 器件 | 通用 | High Thermal Efficiency Tunable Laser with Wide wavelength range and Frequency-Sweep Bandwidth | 级联相移器提升热效率的MCI可调激光器，调谐93 nm、最小线宽48.5 kHz，用于通信与LiDAR |
| W3E.3 | 口头 | 器件 | 通用 | Athermally Fixed-Wavelength Operation of Twin Electro-Optically Tunable RTF Laser for Cost-Effective Scale-Out in Coherent Network | 孪生电光可调RTF激光器20–70°C无跳模无热定波长工作，低成本相干Scale-out |
| W3F.1 | 特邀 | 算法 | 通用 | Optical arbitrary waveform generation using spectro-temporal unitary transforms | 特邀：级联相位调制器+色散元件的谱-时酉变换实现光任意波形生成，不受调制器带宽限制且理论无损 |
光放与多波段
30 篇 · EDFA/BDFA/拉曼/参量放大、S/U/E/O 多波段
| 编号 | 类型 | 技术层 | 场景 | 题目 | 中文要点 |
|---|---|---|---|---|---|
| M2C.5 | 高分 | 光系统 | 长途 | Real-time C+L band Transmission of 64 Tb/s with 1 Tb/s Single-carrier Channels over 1260 km | 超低损大有效面积光纤+混合拉曼/EDFA，实时C+L 64×1 Tb/s单载波1260 km，跨段最长160 km |
| M2E.1 | 特邀 | 器件 | 通用 | High-power single-frequency multimode fiber amplifier with suppressed nonlinearity and focused output | 输入波前整形的单频多模光纤放大器，503 W、斜率效率82%、抑制非线性且聚焦输出 |
| M2E.2 | 口头 | 算法 | 通用 | Toward High-Speed Channel-Specific Gain Shaping: Modelling and Control of an EDFA with a Built-in Dynamic Gain Equalizer | 物理嵌入神经网络对内置动态增益均衡器的EDFA一次性按通道增益整形，最大误差0.3 dB |
| M3C.2 | 高分 | 光系统 | 通用 | 135.5 Tb/s O-Band + 201.4 Tb/s SCL-Band Bidirectional WDM Transmission over a Field Deployed G.652.D Fibre System | 现网G.652.D光纤39 km双向传输：O波段135.51 Tb/s+SCL波段201.39 Tb/s（纪录），掺杂光纤放大 |
| M3C.3 | 高分 | 光系统 | 长途 | Ultra-Wideband Transmission Systems From an Energy Perspective: Which Band is Next? | 从能耗角度看超宽带：1000 km OESCL系统吞吐为CL的2.98倍，每比特能耗高48% |
| M3C.4 | 口头 | 光系统 | 通用 | Design Consideration of Lumped Raman Amplifiers in S-Band Multi-Span Transmission to Minimize Transmission Performance Impairments | S波段集中式拉曼放大器多跨传输损伤分析与考虑色散的设计准则 |
| M3E.2 | 特邀 | 器件 | 海缆 | Multicore Amplifiers for Submarine Networks | 特邀：面向海缆的多芯放大器——受限电功率与安装空间下的挑战 |
| M3E.3 | 高分 | 器件 | 海缆 | Multicore Native Bidirectional 4-core multicore EDFA with Back Reflection Circuit for Submarine Repeaters | 海缆中继器用多芯原生双向4芯EDFA，含背反射电路，免光纤扇入扇出提升空间效率 |
| M3E.5 | 口头 | 器件 | 通用 | Improved Power Efficiency in Cladding-pumped Multi-core Amplifier by Distributed Nonuniform Bend Control | 分布式非均匀弯曲控制提升包层泵浦4芯放大器功率转换效率至18.2%（纪录） |
| M3E.6 | 口头 | 器件 | 通用 | Exceeding 10% Pump-conversion Efficiency for Weakly Coupled Four-core EDFA with Optimized Cladding Profile | 深中心下陷包层结构的弱耦合4芯EDF，C波段功率转换效率10.2%（同类最高） |
| M4C.2 | 特邀 | 器件 | 通用 | Multi-band Raman amplification: progress and challenges | 特邀：多波段拉曼放大的进展与挑战 |
| M4C.3 | 口头 | 器件 | 通用 | Two-Pump Distributed Parametric Amplification of Two 50Gbps PAM4 Channels in SSMF in O-Band | 首次在O波段SSMF中双泵浦分布式参量放大两路50G PAM4，增益16 dB、带宽19 nm、每泵仅~155 mW |
| M4C.4 | 口头 | 算法 | 通用 | ILP-Based Joint Signal–Pump Power Optimization for Hybrid Raman/DFA Amplified Multi-Band Fiber Systems | ILP联合信号-泵浦功率优化混合拉曼/DFA多波段系统，容量310 Tb/s、GSNR平坦度改善2.4 dB |
| M4E.3 | 口头 | 器件 | 通用 | Raman Gain Profiling and Distributed Raman Amplification in a Field-Deployed 15-Mode Graded-Index Multimode Fiber | 现网15模渐变多模光纤拉曼增益剖面与分布式拉曼放大，泵高阶模时增益近均匀 |
| M4K.4 | 口头 | 算法 | 海缆 | Modeling and Experimental Validation of Unrepeated Optical Transmission System with Bi-direction Raman Amplifier for 4D PS-QAM | 260 km双向拉曼无中继系统中4D PS-QAM建模与实验，仿真与实验差距<0.2 dB |
| Th2A.20 | 海报 | 器件 | 通用 | High-efficiency cladding-pumped C+L-band uncoupled 4-core EDFA with integrated multicore isolators and FWDMs | 集成多芯隔离器与FWDM的包层泵浦C+L非耦合4芯EDFA，PCE 9.18% |
| Th4A.8 | PDP | 器件 | 通用 | Wavelength-on-Demand Multimode and Multicore Fiber Optical Parametric Oscillators | 首个无种子多模/多芯光纤参量振荡器平台，泵浦空间态编码波长选择，失谐±45 THz、单线转换40% |
| Th4B.5 | PDP | 光系统 | 城域 | 450 Tb/s GMI, 42.4 THz, OESCL-Band Transmission Over a Field-Deployed Fiber | 39 km现网G.652.D光纤O/E/S/C/L 1273通道42.4 THz最宽带宽传输，吞吐>450 Tb/s(GMI)（SMF纪录） |
| Tu3D.1 | 特邀 | 器件 | 通用 | What’s Next for Doped-Fiber Amplifiers? New Materials, Spectral Bands, and Applications | 特邀：掺杂光纤放大器的下一步——新材料、新波段、高功率及用于空芯光纤与PIC |
| Tu3D.2 | 口头 | 器件 | 通用 | High gain, low-noise broadband hybrid fiber amplifier for L+U band amplification | 光子晶体光纤+掺铋光纤紧凑宽带混合放大器，首次覆盖L+U波段，净增益34.4 dB、NF 3.34 dB |
| Tu3D.3 | 口头 | 器件 | 通用 | Hybrid BiEr Fiber Amplifier for WDM Seamless Amplification of Sand C-Transmission Bands | 铋铒共掺光纤放大器(BiEr-DFA)无缝放大S+C(1490–1560 nm, 9 THz)，增益>20 dB、NF<6 dB、PCE 20% |
| Tu3D.4 | 口头 | 器件 | 通用 | Shortest Fiber Length (<26 m) BDF/EDF Hybrid Amplifier Achieving >20 dB Gain with High Output Power over 166 nm-Wide E+S+C Band | 仅26 m光纤的BDF/EDF混合放大器覆盖E+S+C(1400–1566 nm, 166 nm)，增益20 dB、输出27 dBm |
| Tu3D.5 | 口头 | 器件 | 通用 | Relative Intensity Noise Transfer in O-band Bismuth Doped Fibre Amplifiers | O波段掺铋放大器前向泵浦(Yb激光)存在泵浦-信号RIN转移劣化SNR，后向泵浦可抑制 |
| W1C.2 | 高分 | 光系统 | 长途 | High-Symbol-Rate 192-GBaud Signal Transmission in S+C+L Band over 2000 km with Net Bitrate of 102.8 Tb/s | 低噪前向拉曼泵浦，S+C+L 18.4 THz 92×192 GBd 2000 km，平均单波1.117 Tb/s（最高），净102.8 Tb/s |
| W1H.4 | 口头 | 网络 | 长途 | A Network-Wide Power Analysis of Cladding-Pumped Amplification in Multi-Core Optical Networks | 多芯光网络包层泵浦放大全网功耗分析，每比特功耗较多纤/纤芯泵浦方案最多省37% |
| W2A.19 | 海报 | 器件 | 海缆 | Pumping Power Reduction in Three-stage Tandem Multiband EDFA for Multicore Bidirectional Transmission | ASE再循环使三级串联多波段多芯EDFA泵浦功率降11%，用于海缆多芯双向传输 |
| W2A.22 | 海报 | 器件 | 通用 | Experimental demonstration of L-band discrete Raman amplifier using germanium doped photonic crystal fiber with high Raman gain coefficient and low loss | 掺锗光子晶体光纤L波段(1575–1610 nm)离散拉曼放大器，平均增益25.88 dB、NF<4.78 dB |
| W2A.53 | 海报 | 光系统 | 通用 | 80 Tb/s Real-time Demonstration by an E2E-unified C+L-Band System Across a Seamless 102 nm Bandwidth | 单级EBDFA端到端统一C+L系统，102 nm无缝带宽实时82通道80 km，总80 Tb/s（多数1 Tb/s/通道） |
| W3F.3 | 口头 | 器件 | 通用 | Crosstalk Reduction in Bidirectional Transceiver Using Phase-Sensitive Parametric Amplifier | PPLN相敏参量放大的双向收发机架构，选择性相敏去放大抑制串扰14 dB |
| W3J.4 | 高分 | 光系统 | 通用 | 153.8 Tb/s O-Band Coherent Transmission over SMF with Low-Complexity DSP | O波段16.5 THz WDM相干传输153.8 Tb/s@50.3 km（纪录），几何整形256QAM、免色散补偿、7抽头均衡 |
SDM光纤
22 篇 · 多芯/少模/OAM 光纤与 SDM 传输
| 编号 | 类型 | 技术层 | 场景 | 题目 | 中文要点 |
|---|---|---|---|---|---|
| M2E.6 | 特邀 | 器件 | 通用 | Harnessing Spatial and Dispersion Diversity: Multicore and Few-Mode Fibers for Advanced Signal Processing | 特邀：色散分集多芯/少模SDM光纤用于可调光/微波信号处理（分布式光子处理器） |
| M3E.4 | 口头 | 器件 | 通用 | Low-loss and broadband MDL compensation using LPFG with higher-order cladding-mode coupling | 飞秒刻写长周期光纤光栅(高阶包层模耦合)实现低损宽带MDL补偿，DMA从3.2降至0.17 dB |
| M4E.1 | 特邀 | 器件 | 通用 | Few-Mode Fibers: Why Do They Still Matter? | 特邀：少模光纤为何仍然重要——密度、容量、成本与成缆部署 |
| M4E.2 | 口头 | 器件 | 通用 | S, C and L-Band Characterization of a Differential Mode Delay Compensated Few-Mode Fiber Link | 三模光纤链路在S/C/L(1480–1620 nm)的差分模延迟补偿特性测量 |
| M4E.4 | 口头 | 器件 | 通用 | Experimental characterization of a weakly-guiding ring-core fiber that supports 24 OAM modes including radial high-order modes with low loss and crosstalk | 弱导三环环芯光纤支持24个OAM模（含径向高阶模），低损低串扰 |
| M4E.5 | 口头 | 器件 | 通用 | Experimental demonstration of ultra-long (100 km) ring-core fiber mode transfer matrix characterization and sensing | 自参考共路方法表征100 km环芯光纤模式传递矩阵，并用于远程转速测量 |
| Th1D.4 | 口头 | 器件 | 通用 | Low Loss and Mass Producible Multi-Core Fiber Fan-In/Fan-Out Device Based on Laser Direct Writting Glass Waveguides | 飞秒激光直写玻璃3D波导的4芯光纤扇入扇出，1310 nm平均插损~0.26 dB，写速>30 mm/s可量产 |
| Th2A.17 | 海报 | 器件 | 通用 | Data-Driven Estimation of Nonlinear Factors in Strongly Coupled Multicore Fibers Using Physics-Informed Machine Learning | 物理信息ML数据驱动估计强耦合多芯光纤非线性因子 |
| Th2A.18 | 海报 | 器件 | 长途 | Impact of Core Permutation on the Splice Losses for a Marker-less 7-Core MCF | 无标记60°旋转对称7芯光纤（单预制棒2000 km），纤芯置换熔接代价<0.16 dB，可免芯识别 |
| Th3A.1 | 口头 | 算法 | 通用 | Experimental Demonstration of Non-Data-Aided MIMO Equalization for Strongly Coupled SDM Transmission | 子空间约束初始化CMA+时间正交化实现耦合芯MCF无奇异非数据辅助MIMO均衡，常规CMA 180例全部失败 |
| Th3A.2 | 高分 | 光系统 | 通用 | Real-time 2.5-Pb/s Bidirectional Transmission over 24-core Single-Mode Fiber in S+C+L Bands | 商用400G相干转发器实时2.5 Pb/s双向传输：10.3 km 24芯光纤S+C+L，6288个SDM/WDM通道 |
| Th3A.4 | 口头 | 算法 | 长途 | QR-Decomposition Based Recursive Least Squares Low-Complexity MDL-Tolerant MIMO Equalization for High-Symbol-Rate Coupled SDM Channel Transmissions | QR分解RLS低复杂度MDL容忍MIMO均衡，复杂度降67%，12耦合芯140 GBd 3317 km |
| Th4C.5 | PDP | 光系统 | 海缆 | First Field-Validated Orbital Angular Momentum-Based Submarine Communication | 全球首个OAM-SDM海缆现场试验（南海）：抗氢损低串扰OAM光缆+均匀增益OAM放大器，2232通道 |
| Tu2B.3 | 口头 | 器件 | 通用 | Cost Estimation of Multi-Core Few-Mode Fibers | 多芯少模光纤每芯/每模成本估算：2芯或≤4芯且每芯≥10模时不高于标准单模光纤 |
| Tu3E.5 | 口头 | 光系统 | 通用 | First Real-time Field Demonstration of Space-division Multiplexing Transmission Enabled by 1.2Tb/s/λ OTN and Multi-core Amplifier | 首个多芯放大器支撑的现网SDM实时传输：48 km 4芯光缆单跨153.6 Tb/s，1.2 Tb/s/λ OTN |
| W1F.1 | 口头 | 器件 | 通用 | Ultra-Low Birefringence in High-Density Trench-Assisted Multi-Core Fibers via Fluorine-Doped Stress Rods | 掺氟应力棒抑制高密度沟槽辅助MCF热致双折射，标准包层8芯30 µm芯距最差拍长≥10 m（媲美SMF） |
| W1F.2 | 口头 | 器件 | 通用 | Accuracy Limits of Crosstalk Measurement Techniques for Weakly Coupled Multi-core Fibers | 弱耦合多芯光纤串扰测量方法（功率计/OTDR/波长分辨）精度极限、动态范围与实用性比较 |
| W1F.3 | 口头 | 器件 | 通用 | Synchronous Propagation Delay Measurement for Few-Mode Multi-Core Fibers Based on S2 Method | 基于S²法同步测量少模多芯光纤DMGD与芯间偏斜，2 km 3模7芯光纤精度≤0.1 ps/m，免分芯器件 |
| W1F.4 | 特邀 | 器件 | 通用 | Distributed Measurement of Polarization and Coupling Characteristics of SDM Fibers | 特邀：SDM光纤偏振与耦合特性分布式测量综述，引入“瑞利种子”概念解释多模结果 |
| W1F.5 | 高分 | 器件 | 海缆 | Applicability of Homogeneous/Heterogeneous Multi-core Fibers for Submarine Cables | 含同质/异质G.654兼容4芯光纤的海缆制造与表征，确认成缆与温度串扰变化下具10000 km传输潜力 |
| W1F.7 | 口头 | 器件 | 长途 | Lisbon Underground Multicore Fiber Ring | 里斯本地铁多芯光纤环测试床：144芯光缆含5种MCF共74根，安装MCF超1900 km |
| W2A.18 | 海报 | 器件 | 通用 | Single-Shot Crosstalk Measurement in Multicore Fibers Using Broadband Sources with Enhanced Dynamic Range | 宽带光源单次测量多芯光纤串扰，无需平均，动态范围覆盖>100 km |
相干DSP与编码
23 篇 · 相干 DSP、MIMO、概率整形、FEC、相噪/定时
| 编号 | 类型 | 技术层 | 场景 | 题目 | 中文要点 |
|---|---|---|---|---|---|
| M3B.4 | 口头 | 算法 | 通用 | Simple Adaptive Scheme for Test-Pattern Generation in LR-FEC Inner Code with Chase-2 Decoding | LR-FEC内码Chase-2译码的自适应测试图样生成，测试图样减少>70%无性能损失 |
| M3K.7 | 口头 | 算法 | 通用 | Experimental Investigation on Information Rate and Achievable Security in Y-00 Cipher Transmission Using 214 (=16,384)-Level Probabilistically Shaped QAM | FPGA原型Y-00密码传输采用加密PS-16384QAM，整形增益0.8 dB，窃听检测失败率达99% |
| Th1B.1 | 特邀 | 算法 | 通用 | Dynamic EEPN Impulse Response Modeling and DSP Compensation for Single-Carrier Coherent Receivers | 均衡增强相噪(EEPN)的闭式动态冲激响应模型，明确本振相噪与CD补偿后失真的联系 |
| Th1B.2 | 口头 | 算法 | 长途 | A Blind Domain-Informed Filter Ensemble for Equalization-Enhanced Phase Noise Mitigation | 领域知识引导的盲滤波器集成补偿EEPN，1.6 Tb/s长途相干链路SNR显著提升 |
| Th1B.3 | 口头 | 算法 | DCI | On the Similarity between Equalization-Enhanced Phase Noise and Equalization-Enhanced Timing Jitter | 推导证明EEPN与均衡增强定时抖动(EETJ)都可视为相位/抖动引起的多径失真，可统一架构补偿 |
| Th1B.4 | 口头 | 算法 | 通用 | Feed-Forward Jitter Correction Using Symbol-Domain Timing Error Detector and Simplified Digital Compensation | 基于符号的前馈定时误差检测器+简化数字补偿，低复杂度校正定时抖动 |
| Th1B.6 | 口头 | 算法 | 通用 | An Accurate and Robust Baud-rate Clock Recovery for Coherent Systems | 高波特率相干系统移位Mueller-Muller波特率时钟恢复，Q因子提升0.5–1.7 dB |
| Th2A.41 | 海报 | 光系统 | 通用 | Comparison of Single Carrier FTN-QAM and PCS-QAM for Amplifier-less Coherent Communication Systems | 无放大短距相干中单载波FTN-QAM vs PCS-QAM：FTN-16QAM功率裕量优0.9 dB |
| Th2A.48 | 海报 | 算法 | 通用 | Nonlinearity-Tolerant Temporal Sign Shaping for 64-QAM Optical Fiber Transmissions | 时间符号整形抑制PS-64QAM非线性干扰，有效SNR增益0.33 dB |
| Th2A.49 | 海报 | 算法 | 通用 | Coherent Optical Transmitter IQ Impairments Estimation in the Digital Subcarrier Multiplexing System Based on the Designed Training Sequence | DSCM中基于成对子载波训练序列联合估计发射端IQ幅相失衡与时延偏斜 |
| W1C.1 | 特邀 | 算法 | 长途 | Coded Modulation Targeting Higher Spectral Efficiency in High-Speed Transmission Systems | 特邀：SPC编码概率整形实现迭代译码提升长途频谱效率，≥100 GBd实验验证 |
| W1C.3 | 口头 | 算法 | 通用 | Low-Complexity Subcarrier-Merged Digital Back-Propagation for High-Baud-Rate Subcarrier-Multiplexing Optical Transmission Systems | 低复杂度子载波合并DBP(SM-DBP)，较SCM-DBP复数乘法减少~74%，性能相当 |
| W1C.4 | 口头 | 算法 | 长途 | Mitigating EEPN-Induced Timing Jitter in High Baud Rate Optical Systems: Experimental Validation and DSP Optimization | 130 GBd系统EEPN致定时抖动实验评估，可变带宽定时估计降RMS抖动40–45%，最差块SNR提升~1 dB |
| W1D.1 | 口头 | 算法 | 通用 | Non-Intrusive Separation and Characterization of Transmitter and Receiver Frequency Responses for Coherent Optical Communication System | 无需改硬件的相干收发机收/发频响非侵入分离表征，补偿后69 GBd OSNR灵敏度改善2 dB |
| W1D.3 | 口头 | 算法 | 通用 | Online Joint and Precise Estimation of Transmitter and Receiver IQ-Skew for Ultra-High Baud-Rate Digital Subcarrier Multiplexing Transmissions | 时间正交频域零自相关序列在线联合估计收发IQ偏斜，100 GBd DP-16QAM DSCM精度<0.1 ps |
| W1D.4 | 口头 | 算法 | 通用 | Experimental Validation of Deep Unfolding for Parameter Optimization in MIMO Adaptive Equalization | 深度展开(deep unfolding)优化MIMO自适应均衡参数的实验验证，大幅降低调参成本 |
| W2A.46 | 海报 | 算法 | 长途 | Low-Complexity Hybrid Neural Compensation for Dispersion and Nonlinearity in Coherent Transmission | 低复杂度混合神经网络替代CDC并补偿非线性，896 Gb/s与16640 km系统SNR增益1.47/2.25 dB |
| W2A.48 | 海报 | 芯片 | 通用 | Real-Time FPGA Prototype of an Optimal Interpolation-Based Clock Recovery Algorithm for MPPM-QPSK | 单FPGA实时MPPM-QPSK最优插值时钟恢复，较前工作提升1 dB |
| W3B.1 | 口头 | 算法 | 城域 | Neural Probabilistic Amplitude Shaping for Nonlinear Fiber Channels | 神经概率幅度整形(联合分布学习)，205 km单跨双偏振64QAM较序列选择SNR增益0.5 dB |
| W3B.3 | 特邀 | 算法 | 通用 | Designing Concatenated RS–BCH Codes With Good Performance-Complexity Trade-offs | 设计RS外码+BCH内码级联码的两条经验法则，兼顾性能与复杂度 |
| W3B.4 | 口头 | 算法 | 城域 | Deterministic-ISI Mitigation and Adaptive Equalization for Bandwidth-Limited Coherent Optical Communications | 确定性ISI抑制辅助自适应均衡，60 GBd信号仅7.5 GHz带宽900 km，Q较CMA提升2.3 dB |
| W3B.5 | 口头 | 算法 | 通用 | Real-time Active Power Management for DSCM Coherent Subcarriers Performance Equalization | 400G收发机子载波级实时数字主动功率管理，均衡DSCM各子载波性能，ROSNR显著改善 |
| W3B.6 | 口头 | 算法 | 长途 | Low-Complexity Adaptive Digital Backpropagation via Clock-Tone Strength in Coherent Optical Systems | 时钟音强度引导的低复杂度自适应DBP，1200 km性能相当而复杂度降43.7–51.7% |
光网络架构与控制
31 篇 · ROADM/OXC、规划、SDN 控制、可插拔组网
| 编号 | 类型 | 技术层 | 场景 | 题目 | 中文要点 |
|---|---|---|---|---|---|
| M2C.1 | 特邀 | 光系统 | 通用 | Next-Generation Coherent Optical Transmission Systems and Practical Optimizations | 特邀：相干光从嵌入式到数字相干可插拔的演进及性能-功耗优化权衡 |
| M2D.5 | 口头 | 芯片 | 通用 | Pixelated NOEM Grating for Wavelength-Selective Switching and Spectral Shaping in ROADM Systems | 像素化纳米光机电(NOEM)光栅实现片上波长选择交换与频谱整形，纳秒重构、>100 nm调谐 |
| M2D.6 | 口头 | 芯片 | 通用 | O-band SiPh 2x2x2λ Polarization Diversity Optical Switch with PDL ≤0.5dB | O波段硅光耦合环2×2×2λ偏振分集WSS，平均PDL 0.41 dB（纪录低） |
| M3M.4 | 口头 | 网络 | 通用 | Open-Source Data for Multi-Domain Network Monitoring and Sensing in Optical and Wireless Networks | 光与无线多域网络监测与传感开源数据集及跨域遥测架构，支撑QoT估计与数字孪生 |
| M3Z.1 | Demo | 网络 | DCI | Demonstration of Customer-Owned and Carrier-Controlled Coherent Pluggables using Ownership-Aware YANG Models | 所有权感知YANG模型：客户自有、运营商控制的相干可插拔在现网IPoWDM中的治理控制 |
| M3Z.16 | Demo | 网络 | 通用 | Exploiting ETSI TeraFlowSDN to Control Innovative Optical Networks | 基于ETSI TeraFlowSDN的光控制器，支持IPoWDM灵活栅格与多粒度场景 |
| Th1D.8 | 口头 | 器件 | 通用 | Scalable freeform fan-in 48 waveguide array in glass | 玻璃内3D多通道波导刻写演示器，1D→2D的48波导扇入阵列，损耗与布线密度领先 |
| Th1G.2 | 口头 | 芯片 | 通用 | Three-Mode Reconfigurable Optical Add-Drop Multiplexer | 模式选择相移器构建三模ROADM基本单元，8种交换状态，每模式40 Gb/s路由 |
| Th2A.34 | 海报 | 网络 | 通用 | Demonstration of Self-Healing in IPoWDM Network with Dual Protection Using MBoSDM and NetDevOps Solutions in Support of 6G | 商用设备+MBoSDM原型的IPoWDM双重保护自愈，开通<1分钟、自愈<30 s |
| Th4B.6 | PDP | 网络 | DCI | Multi-Vendor Interoperable 800G Optical Communications over 1602-km through 800G-OpenROADM, 800-ZR and 800GBASE-2xFR4 Fiber Links | 首次多厂商互通800G：800G-OpenROADM(15×100 km)/800-ZR/800GBASE-2×FR4，共1602 km，首用OpenROADM PCS |
| Th4C.2 | PDP | 网络 | 通用 | Experimental Demonstration of Closed-Loop Waveband Protection in S-C-L-Band IPoBDM Fiber Network using Multi-Granular Optical Nodes and TeraFlowSDN Controllers | TeraFlowSDN控制S-C-L波段IPoBDM网络闭环波带保护，多粒度光节点+偏振分辨OPM+等离子体发射机 |
| Th4C.6 | PDP | 光系统 | 海缆 | Coherent Pluggable Optics Achieve Record 400Gb/s per Wavelength Over a 5,682 km Subsea Cable | 相干可插拔模块在5682 km海缆上实现单波400G（容量×距离纪录） |
| Tu2A.1 | 海报 | 光系统 | DCI | Advanced Technologies in Coherent Pluggables and Use Cases | 特邀：相干可插拔先进技术与DCI/城域/长途应用映射 |
| Tu2B.1 | 特邀 | 器件 | 通用 | Suppression of Rayleigh Scattering Loss in Silica Glass Formed by the Compression of its Melted Phase | 熔融态热压缩石英玻璃(~0.8 GPa)通过拓扑修剪抑制瑞利散射损耗 |
| W1E.2 | 口头 | 器件 | 长途 | Record 3 ms Switching Time Wavelength Selective Switch | FP增强LCoS像素+薄液晶盒WSS，切换时间3 ms（纪录，快两个数量级），支撑<50 ms恢复 |
| W1H.5 | 口头 | 网络 | 通用 | GSNR-aware Spectral Slot and Inter-band Wavelength Converter Assignment in Multi-band Optical Networks | GSNR感知频隙与跨波段波长转换器分配算法，SCL网络有IWC容量较CL高38%（无IWC仅8%） |
| W2A.33 | 海报 | 网络 | 城域 | Model for Designing Cost-Effective DSCM-based Point-to-Multipoint Horseshoe Networks | DSCM点到多点vs单载波相干点到点部署成本解析模型，识别城域-区域马蹄形网络汇聚需求 |
| W2A.35 | 海报 | 网络 | 通用 | Regeneration-aware Provisioning for Multi-band Networks with Band Conversion: A Connectivity Graph Approach | 考虑波段/频谱转换的再生感知多波段网络开通算法，同阻塞率吞吐提升达42.9% |
| W2A.47 | 海报 | 算法 | 通用 | Rate-Adaptive Hierarchical Distribution Matcher for 1.6Tbps+ Coherent Optical Modules | 面向1.6T+相干模块的速率自适应分层分布匹配器(HiDM)，32 bit可调，SNR代价<0.064 dB |
| W3A.1 | 特邀 | 网络 | DCI | The Role of Pluggables vs. Embedded Transponders in Optical Networks | 特邀：可插拔与嵌入式转发器在DC/城域/区域/长途中的角色与权衡 |
| W3A.2 | 口头 | 网络 | 通用 | Performance of Core WDM Networks Equipped with 800Gb/s Coherent Pluggable Interfaces | 量化800G相干可插拔(PCS，优于OpenROADM规范)在德国/印度/北美核心WDM网中的收益 |
| W3A.3 | 口头 | 网络 | 通用 | Impact of Tight Channel Spacing with Fiber Switching in WDM Networks using 800ZR+ Interfaces | 纯光纤交换WDM网络中更紧信道间隔（不受波长路由粒度限制）的800ZR+性能评估 |
| W3A.4 | 口头 | 网络 | 通用 | A Progressive First-hit RWA Framework for Optical Networks with Blocking Optical Cross-connects | 面向阻塞型OXC多纤网络的渐进首次命中RWA框架，运行时间最多降460倍且阻塞率不变 |
| W3A.5 | 口头 | 网络 | DCI | Survivable Hybrid FXC-WXC Architecture for Cost-Effective Core/Regional Optical Networks | 低损低成本高韧性混合FXC-WXC核心/区域网络设计算法，扩大ZR+应用，总成本降14–24% |
| W3A.6 | 高分 | 网络 | 城域 | Low-Loss and Infinitely Scalable Flat Modular OXC for Metro-Scale Optical Networks | 稀疏节点内互连的扁平模块化OXC，低损、可无限扩展，城域网100 GBd DP-16QAM验证 |
| W3A.7 | 口头 | 网络 | 通用 | CDC Network Element for Petabit-Per-Second Randomly-Coupled MCF Networks | 随机耦合MCF支撑的4线路侧C+L CDC网络节点，每线路侧1.19 Pb/s，覆盖上下路/直通/光疏导 |
| W4A.1 | 口头 | 网络 | 长途 | Evaluating Capacity Scaling Strategies for Cost-Efficient Optical Network Evolution | 德国核心网收发机/频谱/空间扩容策略技术经济框架，量化容量、CAPEX与每Tb成本 |
| W4A.2 | 口头 | 网络 | 通用 | Impairment Aware Dynamic Routing and Resource Assignment in Multi-Granular Optical Node Networks | 波长+波带交换多粒度光节点网络的损伤感知动态路由与资源分配 |
| W4A.3 | 口头 | 网络 | 通用 | Network Design with Overlapping Spectrum Allocation in DSCM-Enabled Optical Transport | DSCM光传送网双向重叠频谱分配设计，收发机减少18%、承载业务增加116% |
| W4A.5 | 口头 | 网络 | 通用 | Cost-Efficient Optical Network Capacity Upgrade via ROADM Bypassing | 存量弹性光网络通过熔接暗纤实现ROADM旁路扩容，较多纤/多波段方案节省5.6–17% |
| W4A.6 | 口头 | 网络 | 通用 | Minimal-Hardware Single-WSS ROADM Enabling Contention-Less Optical 1+1 Protection with a Multi-Input-Port N×N WSS Employing Joint Switching | 单WSS极简ROADM（多输入端口N×N WSS联合交换），实现无争用光层1+1环保护 |
光纤与测试
9 篇 · 光纤本体、连接、测量表征
| 编号 | 类型 | 技术层 | 场景 | 题目 | 中文要点 |
|---|---|---|---|---|---|
| M4K.6 | 口头 | 算法 | 通用 | First Experimental Investigation of Koopman-Based Fiber Parameter Identification | 首次实验验证Koopman算子光纤参数辨识（SSMF/NZDSF单跨），反向应用显著提升 |
| Th1D.7 | 高分 | 器件 | 通用 | Towards fiber-like-loss for photonic integration from violet to near-IR | CMOS兼容锗硅酸盐集成电路，458–1550 nm谐振Q>1.8亿，自注入锁定窄线宽激光器 |
| Th2A.55 | 海报 | 光系统 | 通用 | 800G Data and Optical Power Transmission over 5 km Double-Clad Fiber | 5 km双包层光纤800G(107 GBd DP-16QAM)与21.8 W光功率共传，双包层光纤供能距离最长 |
| Tu2B.2 | 口头 | 器件 | 通用 | Evaluating Macrobend Loss Reduction in Trench-Assisted Fibers | 任意形状下陷沟槽辅助光纤弯曲损耗抑制的闭式解析公式，数值与实验验证 |
| Tu2J.4 | 口头 | 芯片 | 通用 | Ultrasmall SCL-band Wavelength Multiplexers Using Mosaic-Based Digital Metamaterial | 首个数字超材料马赛克结构超小SCL波段波长复用器，3.6×12 µm²、插损~3 dB |
| Tu3A.4 | 口头 | 光系统 | 海缆 | On Fiber and System Alternatives for 2-Pb/s Submarine Cables | 2 Pb/s海缆四种光纤/系统方案比较，按目标纤对数求最大跨距评估每比特成本 |
| W1F.6 | 口头 | 器件 | 海缆 | Mass Production of Two-core Fiber and Applicability for Pb/s Submarine Cable | 量产Aeff 112 µm²双芯光纤，损耗0.155 dB/km、串扰可忽略，论证构建1 Pb/s跨洋海缆可行 |
| W2A.17 | 海报 | 器件 | 长途 | A four-group two-core fiber with low inter-group XT and low intra-group SMD in a standard 125-μm cladding | 标准125 µm包层4组双芯光纤，组间串扰<−40 dB/100 km、组内SMD 3.67 ps/√km，降低MIMO复杂度 |
| W2A.20 | 海报 | 器件 | 通用 | 20-dB Gain of Broadband Cr-Doped Crystalline Core Fiber for Mode Matching to Standard Single-Mode Fiber by Novel Edge and Brightness Recognition Growth | 新型边缘与亮度识别生长的9 m掺铬晶体芯光纤，增益20 dB且与SMF模场匹配（最小芯径） |