Software-defined Power Communication Network Routing Control Strategy Based on Graph Convolution Network
-
摘要:
传输时延和数据包丢失率是电力通信业务可靠传输重点关注的问题,该文提出一种面向软件定义电力通信网络的最小路径选择度路由控制策略。结合电力通信网络软件定义网络(SDN)集中控制架构的特点,利用图卷积神经网络构建的链路带宽占用率预测模型(LBOP-GCN)分析下一时刻路径带宽占用率。通过三角模算子(TMO)融合路径的传输时延、当前时刻的路径带宽占用率和下一时刻的路径带宽占用率,计算出从源节点到目的节点间不同传输路径的选择度(Q),然后将Q值最小的路径作为SDN控制器下发的流表项。实验结果表明,该文所提出的路由控制策略能有效减小业务传输时延和数据包丢失率。
Abstract:Transmission delay and packet loss rate are critical issues in reliable transmission of power communication services. A minimum path selection routing control strategy for software-defined power communication networks is proposed. Combining the characteristics of the centralized control structure of the software-defined power communication network, a Link Bandwidth Occupancy Predictive model based on Graph Convolutional Network (LBOP-GCN) is built to analyze the route paths bandwidth occupancy in the next period. The selectivity (Q) of different transmission paths from the source node is calculated to the destination node is calculated by using Triangle Modular Operator (TMO) to fuse the transmission delay of the path, the path bandwidth occupancy at the current moment and the path bandwidth occupancy at the next moment. Then the path with the lowest Q value is used as the flow table of the OpenFlow switch delivered by the Software Defined Network (SDN) controller. Experiments show that the proposed routing control strategy can effectively reduce service transmission delay and packet loss rate.
-
表 1 OpenFlow交换机端口和流表状态参数
端口参数 符号 说明 流表参数 符号 说明 $p_{a,q}^1(t)$ rx_packets 接收的数据包数 $f_a^1(t)$ length 交换机流表容量 $p_{a,q}^2(t)$ tx_packets 转发的数据包数 $f_a^2(t)$ priority 流表项匹配次序 $p_{a,q}^3(t)$ rx_bytes 接收的字节数 $f_a^3(t)$ packet_count 根据流表转发的数据包数 $p_{a,q}^4(t)$ tx_bytes 转发的字节数 $f_a^4(t)$ byte_count 根据流表转发的字节数 $p_{a,q}^5(t)$ rx_dropped 接收时丢弃的数据包数 $f_a^5(t)$ duration_sec 数据流持续时间 $p_{a,q}^6(t)$ tx_dropped 转发时丢弃的数据包数 $f_a^6(t)$ duration_nsec 数据流额外生存时间 $p_{a,q}^7(t)$ tx_errors 转发时错误的数据包数 $f_a^7(t)$ idle_timeout 流表项从交换机移除的相对时间 $p_{a,q}^8(t)$ rx_frame_err 接收时错误帧的数 $f_a^8(t)$ hard_timeout 流表项从交换机移除的绝对时间 $p_{a,q}^9(t)$ rx_over_eer 接收时溢出的数据包数 – – – 表 2 链路带宽占用率等级
${\mu _j}(t)$ ${\mu _j}(t)$等级 链路拥塞状态 ${s_j}(t)$ 0~0.6 Ⅰ 无拥塞 1 0.6~0.7 Ⅱ 正常负荷 2 0.7~0.8 Ⅲ 可能拥塞 3 0.8~0.9 Ⅳ 一般拥塞 4 超过0.9 Ⅴ 严重拥塞 5 -
兰巨龙, 于倡和, 胡宇翔, 等. 基于深度增强学习的软件定义网络路由优化机制[J]. 电子与信息学报, 2019, 41(11): 2669–2674. doi: 10.11999/JEIT180870LAN Julong, YU Changhe, HU Yuxiang, et al. SDN routing optimization mechanism based on deep reinforcement learning[J]. Journal of Electronics &Information Technology, 2019, 41(11): 2669–2674. doi: 10.11999/JEIT180870 史久根, 谢熠君, 孙立, 等. 软件定义网络中面向时延和负载的多控制器放置策略[J]. 电子与信息学报, 2019, 41(8): 1869–1876. doi: 10.11999/JEIT181053SHI Jiugen, XIE Yijun, SUN Li, et al. Multi-controller placement strategy based on latency and load in software defined network[J]. Journal of Electronics &Information Technology, 2019, 41(8): 1869–1876. doi: 10.11999/JEIT181053 张红旗, 黄睿, 常德显. 一种基于匹配博弈的服务链协同映射方法[J]. 电子与信息学报, 2019, 41(2): 385–393. doi: 10.11999/JEIT180385ZHANG Hongqi, HUANG Rui, and CHANG Dexian. A collaborative mapping method for service chain based on matching game[J]. Journal of Electronics &Information Technology, 2019, 41(2): 385–393. doi: 10.11999/JEIT180385 于天放, 芮兰兰, 邱雪松. 基于软件定义网络的服务器集群负载均衡技术研究[J]. 电子与信息学报, 2018, 40(12): 3028–3035. doi: 10.11999/JEIT180207YU Tianfang, RUI Lanlan, and QIU Xuesong. Research on SDN-based load balancing technology of server cluster[J]. Journal of Electronics &Information Technology, 2018, 40(12): 3028–3035. doi: 10.11999/JEIT180207 AL-RUBAYE S, KADHUM E, NI Qiang, et al. Industrial internet of things driven by SDN platform for smart grid resiliency[J]. IEEE Internet of Things Journal, 2019, 6(1): 267–277. doi: 10.1109/JIOT.2017.2734903 AUJLA G S, GARG S, BATRA S, et al. DROpS: A demand response optimization scheme in SDN-enabled smart energy ecosystem[J]. Information Sciences, 2019, 476: 453–473. doi: 10.1016/j.ins.2018.09.047 胡宇翔, 李子勇, 胡宗魁, 等. 基于流量工程的软件定义网络控制资源优化机制[J]. 电子与信息学报, 2020, 42(3): 661–668. doi: 10.11999/JEIT190276HU Yuxiang, LI Ziyong, HU Zongkui, et al. Control resource optimization mechanism of SDN based on traffic engineering[J]. Journal of Electronics &Information Technology, 2020, 42(3): 661–668. doi: 10.11999/JEIT190276 熊余, 杨娅娅, 张振振, 等. 软件定义时分波分复用无源光网络中基于带宽预测的资源分配策略[J]. 电子与信息学报, 2019, 41(8): 1885–1892. doi: 10.11999/JEIT180837XIONG Yu, YANG Yaya, ZHANG Zhenzhen, et al. Resource allocation based on bandwidth prediction in software-defined time and wavelength division multiplexed passive optical network[J]. Journal of Electronics &Information Technology, 2019, 41(8): 1885–1892. doi: 10.11999/JEIT180837 伊鹏, 刘洪, 胡宇翔. 一种可扩展的软件定义数据中心网络流调度策略[J]. 电子与信息学报, 2017, 39(4): 825–831. doi: 10.11999/JEIT160623YI Peng, LIU Hong, and HU Yuxiang. A scalable traffic scheduling policy for software defined data center network[J]. Journal of Electronics &Information Technology, 2017, 39(4): 825–831. doi: 10.11999/JEIT160623 王继业, 刘川, 吴军民, 等. 软件定义电力广域网通信业务资源公平分配技术研究[J]. 电网技术, 2015, 39(5): 1425–1431. doi: 10.13335/j.1000-3673.pst.2015.05.038WANG Jiye, LIU Chuan, WU Junmin, et al. Research of software defined service resource equitable allocation technology of power WAN[J]. Power System Technology, 2015, 39(5): 1425–1431. doi: 10.13335/j.1000-3673.pst.2015.05.038 韩宇奇, 何宜倩, 楼凤丹, 等. 基于SDN的动态优化路由策略在信息物理融合电力系统连锁故障中的研究与应用[J]. 电网技术, 2018, 42(8): 2620–2629. doi: 10.13335/j.1000-3673.pst.2017.1796HAN Yuqi, HE Yiqian, LOU Fengdan, et al. Analysis and application of SDN based dynamic optimal route strategy for cyber layer in cascading failures of cyber-physical power system[J]. Power System Technology, 2018, 42(8): 2620–2629. doi: 10.13335/j.1000-3673.pst.2017.1796 TOMOVIC S and RADUSINOVIC I. RO-RO: Routing optimality-reconfiguration overhead balance in software-defined ISP networks[J]. IEEE Journal on Selected Areas in Communications, 2019, 37(5): 997–1011. doi: 10.1109/JSAC.2019.2906762 MANESSI F, ROZZA A, and MANZO M. Dynamic graph convolutional networks[J]. Pattern Recognition, 2020, 97: 107000. doi: 10.1016/j.patcog.2019.107000 ZHANG Zhengchao, LI Meng, LIN Xi, et al. Multistep speed prediction on traffic networks: A deep learning approach considering spatio-temporal dependencies[J]. Transportation Research Part C: Emerging Technologies, 2019, 105: 297–322. doi: 10.1016/j.trc.2019.05.039 PARISOT S, KTENA S I, FERRANTE E, et al. Disease prediction using graph convolutional networks: Application to autism spectrum disorder and Alzheimer’s disease[J]. Medical Image Analysis, 2018, 48: 117–130. doi: 10.1016/j.media.2018.06.001 贾惠彬, 薛凯夫, 马静, 等. 广域保护通信多路径路由选择的改进蚁群算法[J]. 电力系统自动化, 2016, 40(22): 22–26. doi: 10.7500/AEPS20160612005JIA Huibin, XUE Kaifu, MA Jing, et al. Improved ant colony algorithm for multi-path routing selection in wide-area protection communication system[J]. Automation of Electric Power Systems, 2016, 40(22): 22–26. doi: 10.7500/AEPS20160612005 SUN Yi, YU Li, ZHANG Jing, et al. Joint MAC-PHY layer resource allocation algorithm based on triangle module operator for multi-service OFDM system[J]. Procedia Environmental Sciences, 2011, 10: 163–169. doi: 10.1016/j.proenv.2011.09.029 WANG Junfeng, MIAO Yiming, ZHOU Ping, et al. A software defined network routing in wireless multihop network[J]. Journal of Network and Computer Applications, 2017, 85: 76–83. doi: 10.1016/j.jnca.2016.12.007 AI-KASHOASH H A A, AMER H M, MIHAYLOVA L, et al. Optimization-based hybrid congestion alleviation for 6LoWPAN networks[J]. IEEE Internet of Things Journal, 2017, 4(6): 2070–2081. doi: 10.1109/jiot.2017.2754918 樊冰, 唐良瑞. 电力通信网脆弱性分析[J]. 中国电机工程学报, 2014, 34(7): 1191–1197. doi: 10.13334/j.0258-8013.pcsee.2014.07.022FAN Bing and TANG Liangrui. Vulnerability analysis of power communication network[J]. Proceedings of the CSEE, 2014, 34(7): 1191–1197. doi: 10.13334/j.0258-8013.pcsee.2014.07.022