一种面向电力OTN的业务主备路由时延差优化算法
首发时间:2024-03-29
摘要:随着智能电网的发展,传统以波分复用技术为基础的电力通信网络难以满足新型电力通信业务多粒度传输场景。OTN已进入第五代固定网络阶段,可以提供最小带宽约为2Mbps的灵活连接,是下一代电力通信光网络的关键技术。目前OTN中的路由算法不能满足电力通信业务的保护需求和双向时延差需求。针对这一问题,本文提出了一种面向电力OTN的业务主备路由时延差优化算法,该算法以业务的主备路由时延差和全网的风险均衡性为优化目标。在算法实现方面,以KSP算法为基础,通过主备路径最大不相交以及OTN的路由频谱分配等约束对所得业务候选路径集进行筛选,最后依据优化目标函数选择业务的主备路由,在实现全网风险均衡的同时,极大降低了业务主备路由时延差。
For information in English, please click here
Optimization Algorithm for Latency Difference of Service Work and Backup Routes for OTN
Abstract:With the development of smart grids, traditional power communication networks based on wavelength division multiplexing technology are unable to meet the multi granularity transmission scenarios of new power communication services. OTN has entered the fifth generation fixed network stage and can provide flexible connections with a minimum bandwidth of about 2Mbps, making it a key technology for the next generation of power communication optical networks. At present, the routing algorithms in OTN cannot meet the protection needs and bidirectional latency difference requirements of power communication services. In response to this issue, this paper proposes an optimization algorithm for the latency difference between the work and backup routes of a new type of power OTN. The algorithm aims to optimize the latency difference between the work and backup routes of the service and the risk balance of the entire network. In terms of algorithm implementation, based on the KSP algorithm, the candidate path set for the obtained service is filtered through constraints such as maximum non intersection of the work and backup paths and OTN routing spectrum allocation. Finally, the work and backup routes for the service are selected according to the optimization objective function, which greatly reduces the latency difference between the work and backup routes of the service while achieving network risk balance.
Keywords: OTN latency difference of work and backup routes Risk balance KSP algorithm
基金:
引用
No.****
同行评议
勘误表
一种面向电力OTN的业务主备路由时延差优化算法
评论
全部评论0/1000