Sen Shen

LLM-driven Multi-Agent Optical Digital Twin for Automated Data Generation

Sen Shen, Zhuojun Cai, Jing Han, Wanxin Zhao et autres

We propose a multi-agent autonomous framework for optical-network data generation. By integrating LLM-based orchestration with GNPy and OptiCommPy, the system supports automated scenario expansion and iterative physical-layer analysis in a closed loop, accelerating data generation and enabling efficient network optimisation for next-generation …

Delay-Aware Time-Coordinated Nanosecond Optical Switching for Dynamically Reconfigurable Optical Networks

Ali Mehrpooya, Vaigai N Yokar, Yiran Teng, Sen Shen et autres

We demonstrate an FPGA-embedded AI control platform enabling real-time, nanosecond-accurate optical path reconfiguration. Using an on-chip MLP model for delay prediction, our system reduces packet loss by 90%, achieving deterministic, self-healing switching for future AI-native optical networks.

us, gb (code pays fourni par la source)

Bidirectional SOP Sensing for Infrastructure-Aware Optical Networks

Wanxin Zhao, Sen Shen, Yiran Teng, Shuangyi Yan et autres

Optical networks lack direct awareness of external infrastructure conditions along deployed fibre routes. This paper presents a bidirectional SOP-based integrated sensing and communication framework that harvests route-level and eventlevel information from live metropolitan fibre links. Experiments on 7.776 km and 8 km …

LLM-Orchestrated Optical-Fibre ISAC Platform for Sensing-Event Monitoring and Selective Network Response

W Zhao, Sen Shen, Peiyun Ge, Jing Han et autres

We propose an LLM-driven integrated communication-and-sensing system for automated QoT monitoring, sensing-event interpretation, vibration localization, and adaptive network response. Our use case demonstrates real-time vibration-event classification and localization for proactive link-risk assessment.

cn, us (code pays fourni par la source)

Nanosecond-scale photonic switching with dual time synchronization for fast reconfigurable fronthaul in 5G/6G networks

Bohao Sun, Tongyun Li, Heyang Long, Peng Bao et autres

To address the challenges of time-sensitive fronthaul transmission in next-generation radio access networks (RANs), this paper proposes a nanosecond-scale photonic switching architecture enabled by dual precise time synchronization. A fully packaged tri-layered 8×8 electro-optic microdisk optical switch is designed and implemented, driven …

gb, se, us (code pays fourni par la source)

Collaborative Measurement-Driven Digital Twin with Data Extrapolation for DRL-assisted Optical Networks

Yiran Teng, Ruizhi Yang, Sen Shen, Yuelin Liu et autres

We present a measurement-driven digital twin framework that integrates multiple lab-based emulations with AI-based extrapolation, offering a practical approach to data-driven digital twin construction. Using 5% measured data, it achieves below 0.05 dB error and enables DRL-based spectrum allocation with 0.4 dB …

us (code pays fourni par la source)

Delay-Aware Time-Coordinated Nanosecond Optical Switching for Dynamically Reconfigurable Optical Networks

Ali Mehrpooya, Vaigai Nayaki Yokar, Yiran Teng, Sen Shen et autres

We demonstrate an FPGA-embedded AI control platform enabling real-time, nanosecond-accurate optical path reconfiguration. Using an on-chip MLP model for delay prediction, our system reduces packet loss by 90%, achieving deterministic, self-healing switching for future AI-native optical networks.

gb, us (code pays fourni par la source)

A Knowledge-Enhanced Large Language Model Advisor for Degradation-Aware Microgrid Battery Dispatch Under Extreme Heat

Sen Shen, Li Yi, Yi-Xiang Wang, Jiaxu Duan et autres

Battery energy storage dispatch in microgrids involves a difficult tradeoff between operating cost and battery degradation, especially under extreme heat. This paper proposes a knowledge-enhanced large language model advisor for degradation-aware microgrid battery dispatch. The framework combines a knowledge graph (KG) of …

cn, hk (code pays fourni par la source)

Train Flow Sensing and Positioning via In-Service Polarization Reuse over Bidirectional Optical fibre Transmission Systems

Wanxin Zhao, Sen Shen, Yiran Teng, Vaigai Yokar et autres

We propose and deploy an ISIC platform based on polarization and neural-network analysis, and conduct a two-week field trial in a metro network, achieving real-time train traffic monitoring and localization while maintaining high-quality communication.

cn, gb (code pays fourni par la source)

Bidirectional Polarization-Based Integrated Sensing and Communication for In-Service Metropolitan Fibres: Train-Induced Vibration Localization and Train Flow Monitoring

Wanxin Zhao, Sen Shen, Vaigai Nayaki Yokar, Ali Mehrpooya et autres

In-service metropolitan fibre sensing can extend the value of deployed communication links, but practical sensing schemes must remain compatible with live traffic and modest hardware budgets. Here, we present a low-overhead state-ofpolarization (SOP)-based integrated sensing and communication (ISAC) framework that combines simplified …

gb, cn (code pays fourni par la source)

Open-Source Data for Multi-Domain Network Monitoring and Sensing in Optical and Wireless Networks

Sen Shen, Zhou Xueqing, Zhao Wanxin, Yang Ruizhi et autres

We propose a cross-domain telemetry architecture for optical and wireless networks, and collect multi-source datasets, supporting AI engines and network DTs. Thereleased datasets provide insights for QoT estimation, fibre sensing and multi-domain network optimisation.

gb (code pays fourni par la source)

Open-Source Data for Multi-Domain Network Monitoring and Sensing in Optical and Wireless Networks

Sen Shen, Zhou Xueqing, Zhao Wanxin, Yang Ruizhi et autres

We propose a cross-domain telemetry architecture for optical and wireless networks, and collect multi-source datasets, supporting AI engines and network DTs. Thereleased datasets provide insights for QoT estimation, fibre sensing and multi-domain network optimisation.

gb (code pays fourni par la source)