This paper proposes a static Routing and Spectrum Allocation (RSA) method to design low-power elastic optical networks. In the elastic optical networks, data are transmitted via lightpaths, which are routes flexibly assigned frequency resources according to transmission distances and traffic demands. When designing the elastic optical networks with a static lightpath establishment policy, lightpaths are generally established so as to minimize frequency bandwidth allocated in order to save frequency resources necessary for constructing the networks. The proposed method focuses on not only this performance metric but also the power consumption of optical amplifiers. An optical amplifier in an optical fiber works when any of optical signals pass through it, and thus in multifiber environments where each link consists of multiple optical fibers, lightpaths should be collectively amplified from the view point of the power efficiency. Therefore, the relationship between these two performance metrics is a trade-off. The proposed method develops an Integer Linear Programming (ILP) formulation for the RSA problem taking the trade-off into account. Furthermore, the proposed method provides a heuristic algorithm that performs route selection and frequency spectrum allocation for reducing the total power consumption of optical amplifiers while saving necessary frequency resources. Through numerical experiments, we show the performance of the proposed method.

本文提出了一种静态路由和频谱分配（RSA）方法来设计低功率弹性光网络。在弹性光网络中，数据是通过光路传输的，光路是根据传输距离和流量需求灵活分配频率资源的路由。当设计具有静态光路建立策略的弹性光网络时，通常建立光路以最小化分配的频率带宽，以便节省构建网络所需的频率资源。所提出的方法不仅关注该性能指标，而且关注光放大器的功耗。光纤中的光放大器会在任何光信号通过时起作用，因此在多光纤环境中，其中每个链路都由多根光纤组成，从功率效率的角度来看，应将光路集中放大。因此，这两个性能指标之间的关系是一个折衷。所提出的方法针对折衷问题，针对RSA问题开发了整数线性规划（ILP）公式。此外，提出的方法提供了一种启发式算法，该算法执行路由选择和频谱分配，以减少光放大器的总功耗，同时节省必要的频率资源。通过数值实验，我们证明了该方法的性能。所提出的方法针对折衷问题，针对RSA问题开发了整数线性规划（ILP）公式。此外，提出的方法提供了一种启发式算法，该算法执行路由选择和频谱分配，以减少光放大器的总功耗，同时节省必要的频率资源。通过数值实验，我们证明了该方法的性能。所提出的方法针对折衷问题，针对RSA问题开发了整数线性规划（ILP）公式。此外，提出的方法提供了一种启发式算法，该算法执行路由选择和频谱分配，以减少光放大器的总功耗，同时节省必要的频率资源。通过数值实验，我们证明了该方法的性能。