Routing, Restoration and Reconfiguration in Lightwave Networks

本研究的目的是为电路交换波分复用（WDM）网络中的光路路由、恢复和重新配置提供有效的解决方案。这样的网络对下一代互联网的成功至关重要，下一代互联网将有能力以更低的延迟承载比今天的互联网大得多的流量。网络层简单地使用光路作为虚拟链路，在其上执行存储转发分组路由。电子分组传输的光层体系结构的设计对光网络架构师来说是一个巨大的挑战。本课题旨在通过提高光网络架构的容错性和适应性，为光层设计做出贡献。这个项目有两个主要组成部分。第一个部分是开发算法，在链路故障的情况下有效地恢复光路。这些算法不仅可以有效地利用网络资源，即波长和交换机，而且可以提供快速的恢复能力。这需要在资源分配时共同设计服务和恢复路径，从而减少在链路或节点故障时重新路由光路所需的波长数量。研究人员将考虑基于网格的恢复算法，并从波长要求和恢复速度两方面评估其性能。除了确定每个光路的服务和恢复路径外，路由/恢复算法还将指定有效的信号机制，以协调切换到恢复路径和波长分配到这些路径。将分析所提出的信号机制的恢复时间和计算复杂度。基于光路的互联网骨干网通过改变光路的集合为逻辑拓扑的重新配置提供了一个独特的机会。这种重新配置必须基于对逻辑链路上的分组流量的实时测量，并且可以在分组网络的容量规划中提供额外的自由度。研究人员将设计算法，用于确定拓扑重构的频率和性质。在光层引入动态拓扑变换来满足网络层的动态流量需求是一个很有前途的概念，有望开辟许多有趣的研究方向。

The objective of the proposed research is to develop efficient solutions for routing, restoration, and reconfiguration of lightpaths in a circuit-switched wavelength division multiplexed (WDM) network. Such a network is fundamentally important to the success of a next-generation Internet which will have the capacity to carry a much larger amount of traffic than today's Internet at lower delays. The network layer simply uses the lightpaths as the virtual links over which store-and-forward packet routing is performed. The design of the optical layer architecture over which electronic packet transport takes place is a grant challenge for optical network architects. The proposed project aims to contribute to this optical layer design by improving the fault tolerance and adaptability of the optical network architecture. There are two major components to this project. The first component is to develop algorithms that efficiently restore lightpaths in the event of link failures. These algorithms will not only utilize the network resources, namely wavelengths and switches, efficiently, but also provide rapid restoration capability. This requires the design of service and restoration paths jointly at the time of resource allocation, thereby reducing the number of wavelengths required to reroute the lightpaths in the event of a link or node failure. The researchers will consider mesh-based restoration algorithms and evaluate their performance both in terms of wavelength requirement and restoration speed. In addition to determining service and restoration paths for each lightpath, the routing/restoration algorithms will also specify efficient signalling mechanisms to coordinate the switchover to the restoration paths and the allocation of wavelengths to such paths. Restoration times and the computational complexity of the proposed signalling mechanisms will be analyzed. A lightpath-based Internet backbone provides a unique opportunity for the reconfiguration of the logical topology by changing the set of lightpaths. Such reconfiguration must be based on real-time measurement of packet traffic on the logical links and can provide an additional degree of freedom in the capacity planning for packet networks. The researchers will design algorithms that will be used to determine the frequency and the nature of topology reconfiguration. The introduction of dynamic topological transformation at the optical layer to satisfy the dynamic traffic requirements at the network layer is a promising concept that is expected to open many interesting research directions.