Performance Monitoring of Coherent Optical Networks

在过去的几十年中，运营商网络中业务量的持续增长导致了波分复用(WDM)传输系统的发展，其比特率和容量都在不断增加。当前相干系统的总容量高达近10 TB/S(96通道x 100G/通道)，这使得链路和网络级别的性能下降和故障的潜在影响比以往任何时候都更加严重，对监控的需求比以往任何时候都更加重要。该研究项目旨在基于在运营商网络中收集的多个季节和多个光纤工厂部署的真实监测数据，开发大容量相干网络的性能降级和故障预测模型。该项目建立在首席研究员在光纤测量和网络领域开展的研究活动的基础上。这一探索性研究的结果将有助于在网络层面制定预防策略，使大容量相干网络更可靠、更易于网络运营商管理，并为未来的光纤网络部署制定更精确的工程和规划规则。其他创新包括：相干网络中的长期性能动态；不同外部设备中性能演变和极化活动的新链路度量、时间常数和预测方法；由相干应答器的数字信号处理(DSP)能力实现的新的主动式故障管理应用；新的软件定义的网络(SDN)应用，因为可以预期主动式和自动化故障管理方案允许通过简化和自动化的网络操作和更高的网络可靠性来开发新的应用。该研究项目将与Ciena以及Verizon和KTH皇家理工学院的合作者合作实现。为期3年的研究项目将为学生提供一个独特的机会，培训他们关于光纤性能监控和先进的网络管理概念和实践，并将培训至少2名博士和5名硕士学生。**

The continuous traffic increase in carrier networks in the last decades has led to the development of wavelength division multiplexing (WDM) transmission systems with ever-increasing bitrates and capacity. Current coherent systems exhibit a total capacity of up to nearly 10 Tb/s (96 channels x 100G/channel), which makes the potential impact of performance degradation and failure at both the link and network levels more severe and the need for monitoring more important than ever. This research project aims at developing predictive models for performance degradation and failure for high-capacity coherent networks based on real monitoring data collected in a carrier network over several seasons and on multiple fiber plant deployments. This project builds on the research activities conducted by the principal investigator in the area of optical fiber measurements and networks. The results of this exploratory research will enable to develop preventive strategies at the network level for making high-capacity coherent networks more reliable and easier to manage for network operators, as well as more precise engineering and planning rules for future fiber network deployments. Some additional innovations are: long-term performance dynamics in coherent networks; new link metrics, time constants and predictive methods for performance evolution and polarization activity in different outside plants; new proactive fault management applications enabled by the digital signal processing (DSP) capabilities of coherent transponders; new software defined network (SDN) applications, given that proactive and automated fault management schemes can be expected to allow for the development of new applications through simplified and automated network operations and higher network reliability. The research project will be realized in partnership with Ciena, as well as collaborators from Verizon and KTH Royal Institute of Technology. The 3-year research project will provide a unique opportunity for training students on optical performance monitoring and advanced network management concepts and practices and will allow training at least 2 Ph.D. and 5 Master students.**