应用于物联网边缘网关的队列调度算法研究  

Research on Queue Scheduling Algorithm Applied to IoT Edge Gateway

作　　者：袁瑞 蒋伟 胡静[1] 宋铁成[1] YUAN Rui;JIANG Wei;HU Jing;SONG Tie-cheng(School of Information Science and Engineering,Southeast University,Nanjing 211102,China)

机构地区：[1]东南大学信息科学与工程学院,江苏南京211102

出　　处：《测控技术》2022年第12期72-77,共6页Measurement & Control Technology

基　　金：国家重点研发计划(2020YFB1600104);江苏省重点研发计划(BE2020084-2)

摘　　要：物联网的发展对网关传输能力提出了更高的QoS需求。现有的队列调度算法在链路拥堵时不能为实时数据提供及时的转发服务,因此在WF2Q+算法的基础上引入实时队列和低时延队列的概念。利用物联网网关的边缘计算能力,基于改进的WF2Q+算法建立了单网关优化传输模型,同时提出多网关协同传输模型。仿真证明,改进的WF2Q+算法可以很好地在网络拥塞情况下降低实时队列和低时延队列的传输时延,级联令牌桶保证了突发数据流下网关传输模型的稳定性,多网关协同传输模型可以使数据自主选择传输路径,降低整体传输时延。The development of Internet of things puts forward higher QoS capability requirements for gateway transmission.When congestion occurs,existing queue scheduling algorithms cannot provide timely forwarding service for realtime data.Therefore,the concepts of realtime queue and low latency queue are introduced based on WF2Q+algorithm.Considering the edge computing capability of the Internet of things gateway,a singlegateway optimized transmission model which based on the improved WF2Q+algorithm is established,and a multigateway cooperative transmission model is proposed.Simulation results show that the improved WF2Q+algorithm can effectively reduce the transmission delay of realtime queue and low latency queue under the condition of network congestion,with token bucket algorithm ensuring the stability of the transmission model under burst data stream.Meanwhile,the multigateway transmission model can let the data choose the transmission path autonomously,thus reducing the overall transmission delay.

关 键 词：队列调度 协同传输 WF2Q+ 低时延队列 令牌桶算法 

分 类 号：TN919[电子电信—通信与信息系统] TP393[电子电信—信息与通信工程]