摘要
能源互联网新业务的快速增长对弹性光网络的业务流识别技术提出了更为复杂的要求。为了提高弹性光网络对多业务的匹配支撑能力,本文首先对弹性光网络的业务流识别需求进行了分析,提出了一种轻量化回声状态网络模型。该模型对传统的回声状态网络模型进行了裁剪,使储备池的神经元结构更为简单。针对弹性光网络业务流识别需求,提出了基于轻量化回声状态网络的业务流识别优化算法,并设计其在弹性光网络的实现方案。本文以能源互联网多业务混合接入作为典型场景,对弹性光网络的轻量化业务识别机制搭建仿真软件进行验证。验证结果分析表明本文所提出的算法能较好地适用于光网络系统,有效提高其对多业务接入的QoS保障。
The rapid growth of new services on the energy Internet puts forward more complex requirements for the service flow identification technology of elastic optical networks.In order to improve the matching support ability of elastic optical networks for multiple services,this paper first analyzes the service flow identification requirements of elastic optical networks,and proposes a lightweight echo state network model.This model tailors the traditional echo state network model,so that the neuronal structure of the reserve pool becomes simpler.In order to meet the requirements of service flow identification in elastic optical networks,a service flow identification optimization algorithm based on lightweight echo state network is proposed,and its implementation scheme in elastic optical networks is designed.In this paper,the multiservice hybrid access of the energy Internet is taken as a typical scenario to verify the simulation software established for the lightweight service identification mechanism of the elastic optical network.The analysis of the verification results shows that the proposed algorithm can be well applied to the optical network system,and effectively improve its QoS guarantee for multi-service access.
作者
吉晓龙
静国刚
Ji Xiao-long;Jing Guo-gang(The Third Research Institute of the Ministry of Public Security,Shanghai 200031,China)
出处
《科学与信息化》
2023年第23期76-80,共5页
Technology and Information
关键词
轻量化回声状态网络
弹性光网络
业务流识别
能源互联网
lightweight echo state network
elastic optical network
service flow identification
energy Internet
