摘要
弹性光网络(EON)中的传统路由频谱分配(RSA)算法多考虑路由跳数或频谱资源占用情况,缺乏时域与相邻链路的信息有效利用。提出一种结合预测的多维感知RSA算法,对持续时间已知业务的历史时间信息通过后向传播神经网络预测未来业务的时间信息,在路由时综合考虑时间、频谱和相邻链路资源占用程度。仿真结果表明:与传统RSA算法相比,多维感知RSA算法能有效降低带宽阻塞率。
The traditional routing spectrum allocation(RSA)algorithm in elastic optical network(EON)mostly considers the hop number or spectrum resource occupation,and lacks the effective utilization of time domain and adjacent link information.This paper proposes a multi-dimension-aware RSA algorithm combined with prediction,which predicts the time information of future services through back propagation neural network for the known duration of services time,spectrum and adjacent link resource occupancy are considered.The simulation results show that compared with the traditional RSA algorithm,multi-dimension-aware RSA algorithm can effectively reduce the bandwidth blocking rate.
作者
徐凯
宣涵
陆煜斌
王炎豪
朱嘉豪
沈建华
XU Kai;XUAN Han;LU Yubin;WANG Yanhao;ZHU Jiahao;SHEN Jianhua(School of Communications and Information Engineering,Nanjing University of Posts andTelecommunications(NJUPT),Nanjing 210003,China)
出处
《光通信技术》
2021年第4期43-47,共5页
Optical Communication Technology
关键词
弹性光网络
路由和频谱分配
后向传播神经网络
多维感知
elastic optical networks
routing and spectrum allocation
back propagation neural networks
multi-dimension-aware
