Over the past few decades, the world has witnessed a rapid growth in mobile and wireless networks（MWNs） which significantly change human life. However, proliferating mobile demands lead to several intractable challe...Over the past few decades, the world has witnessed a rapid growth in mobile and wireless networks（MWNs） which significantly change human life. However, proliferating mobile demands lead to several intractable challenges that MWN has to face. Software-defined network is expected as a promising way for future network and has captured growing attention. Network virtualization is an essential feature in software-defined wireless network（SDWN）, and it brings two new entities, physical networks and virtual networks. Accordingly, efficiently assigning spectrum resource to virtual networks is one of the fundamental problems in SDWN. Directly orienting towards the spectrum resource allocation problem, firstly, the fluctuation features of virtual network requirements in SDWN are researched, and the opportunistic spectrum sharing method is introduced to SDWN. Then, the problem is proved as NP-hardness. After that, a dynamic programming and graph theory based spectrum sharing algorithm is proposed.Simulations demonstrate that the opportunistic spectrum sharing method conspicuously improves the system performance up to around 20%–30% in SDWN, and the proposed algorithm achieves more efficient performance.展开更多
Large latency of applications will bring revenue loss to cloud infrastructure providers in the cloud data center. The existing controllers of software-defined networking architecture can fetch and process traffic info...Large latency of applications will bring revenue loss to cloud infrastructure providers in the cloud data center. The existing controllers of software-defined networking architecture can fetch and process traffic information in the network. Therefore, the controllers can only optimize the network latency of applications. However, the serving latency of applications is also an important factor in delivered user-experience for arrival requests. Unintelligent request routing will cause large serving latency if arrival requests are allocated to overloaded virtual machines. To deal with the request routing problem, this paper proposes the workload-aware software-defined networking controller architecture. Then, request routing algorithms are proposed to minimize the total round trip time for every type of request by considering the congestion in the network and the workload in virtual machines（VMs）. This paper finally provides the evaluation of the proposed algorithms in a simulated prototype. The simulation results show that the proposed methodology is efficient compared with the existing approaches.展开更多
针对传统的IP欺骗攻击缓解方法存在运算开销大、缺乏灵活性等问题,提出了一种基于动态限制策略的软件定义网络(software defined network,SDN)中IP欺骗攻击缓解方法。首先,利用Packet-In消息中三元组信息回溯攻击路径,定位IP欺骗攻击源...针对传统的IP欺骗攻击缓解方法存在运算开销大、缺乏灵活性等问题,提出了一种基于动态限制策略的软件定义网络(software defined network,SDN)中IP欺骗攻击缓解方法。首先,利用Packet-In消息中三元组信息回溯攻击路径,定位IP欺骗攻击源头主机;然后,由控制器制定动态限制策略对连接攻击源头主机的交换机端口的新流转发功能进行限制,待限制期满再恢复其转发新流的功能,限制期的大小随着被检测为攻击源的次数而增长。研究结果表明:这种动态的限制策略可阻隔攻击流进入SDN网络,从而有效避免SDN交换机、控制器以及链路过载;由于在限制期间无需再对这些限制的交换机端口进行实时监测,该方法在应对长时攻击时较传统方法具有更高的缓解效率和更少的资源消耗。展开更多
针对工业物联网中业务需求多样性和服务质量(Quality of Service,QoS)要求差异性导致的网络资源利用低问题,提出一种基于深度强化学习的网络切片资源分配策略。该策略运用深度强化学习优化网络切片资源分配的准入控制,通过智能体在特定...针对工业物联网中业务需求多样性和服务质量(Quality of Service,QoS)要求差异性导致的网络资源利用低问题,提出一种基于深度强化学习的网络切片资源分配策略。该策略运用深度强化学习优化网络切片资源分配的准入控制,通过智能体在特定时间窗口内处理资源请求,并根据不同网络切片的QoS要求及请求准入结果进行资源的动态分配。实验结果表明,所提策略相比基准算法在提高网络收益、资源利用率和接收率方面分别提升了8.33%、9.84%和8.57%。该策略能够在保证服务质量的同时提高整个网络的效率和性能。展开更多
软件定义网络(Software Defined Network,SDN)是一种新型的网络架构,相比于传统网络,简化了网络管理并更好地支持网络流量的动态控制,现已被许多应用领域采用。为了增强SDN的异常检测与防御能力,在拜占庭容错机制的基础上,提出一种异常...软件定义网络(Software Defined Network,SDN)是一种新型的网络架构,相比于传统网络,简化了网络管理并更好地支持网络流量的动态控制,现已被许多应用领域采用。为了增强SDN的异常检测与防御能力,在拜占庭容错机制的基础上,提出一种异常检测方法,对异常或错误的指令容错,保证正确下发流表,同时通过理论分析证明该方法的有效性和安全性。实验表明,在SDN网络环境中,该检测方法能够快速检测出异常网络设备,降低SDN异常检测中的漏报率和误报率。展开更多
针对SDN流量工程中流量预测基于静态时空依赖的问题,提出了一种基于注意力机制的图卷积神经网络(GCN)与门控递归单元(GRU)集成的动态网络流量预测方法——AGCNGRU(attention mechanism for GCNGRU model)。借助GCN捕获网络中节点之间的...针对SDN流量工程中流量预测基于静态时空依赖的问题,提出了一种基于注意力机制的图卷积神经网络(GCN)与门控递归单元(GRU)集成的动态网络流量预测方法——AGCNGRU(attention mechanism for GCNGRU model)。借助GCN捕获网络中节点之间的流量空间依赖性和GRU捕获流量经过网络中各节点的时间依赖性,通过时间注意力机制设计每个隐藏状态的权重,以调整时间点流量信息的重要性,同时通过数据驱动空间注意力机制动态自适应调整Laplace矩阵,实现动态提取网络信息数据时空相关性,最终完成动态流量精准预测。在GEANT的数据集上的实验表明,所提出的方法在均方误差方面比GCNGRU减少24.8%,比GRU减少66.4%,并通过与传统路由算法OSPF、DDPG算法比较,在90%的流量负载强度下,网络性能比OSPF提升了24%,比DDPG提升了8.1%,进一步说明了AGCNGRU算法网络流量准确预测带来的时效性和有效性。展开更多
随着智能电网的快速发展,配电网中信息物理耦合关系日益紧密。这种耦合性使得配电网更容易被多方面极端事件所影响,在通信网络发生故障时会降低系统的态势感知和控制能力,从而制约配电网的灾后负荷恢复能力,因此通信网络恢复对灾后配电...随着智能电网的快速发展,配电网中信息物理耦合关系日益紧密。这种耦合性使得配电网更容易被多方面极端事件所影响,在通信网络发生故障时会降低系统的态势感知和控制能力,从而制约配电网的灾后负荷恢复能力,因此通信网络恢复对灾后配电网负荷恢复至关重要。该文提出一种通信网络恢复和负荷恢复的协同优化决策方案,该方案将环网通信网络与软件定义网络(software defined networking,SDN)技术相结合,灵活恢复灾后的配电网通信网络,进而控制配电网拓扑重构形成以分布式电源为中心的微电网以恢复负荷电力供应,并进一步使用一种信息物理协同的启发式计算方法实现恢复方案的快速计算。最后,使用IEEE 33节点和IEEE 123节点测试系统验证所提出方法的优点和有效性。展开更多
基金supported by the National Natural Science Foundation of China(6102100161133015+4 种基金61171065)the National Natural Science Foundation of China(973 Program)(2013CB329001)the National High Technology ResearchDevelopment Program(863 Program)(2013AA0106052013AA013500)
文摘Over the past few decades, the world has witnessed a rapid growth in mobile and wireless networks（MWNs） which significantly change human life. However, proliferating mobile demands lead to several intractable challenges that MWN has to face. Software-defined network is expected as a promising way for future network and has captured growing attention. Network virtualization is an essential feature in software-defined wireless network（SDWN）, and it brings two new entities, physical networks and virtual networks. Accordingly, efficiently assigning spectrum resource to virtual networks is one of the fundamental problems in SDWN. Directly orienting towards the spectrum resource allocation problem, firstly, the fluctuation features of virtual network requirements in SDWN are researched, and the opportunistic spectrum sharing method is introduced to SDWN. Then, the problem is proved as NP-hardness. After that, a dynamic programming and graph theory based spectrum sharing algorithm is proposed.Simulations demonstrate that the opportunistic spectrum sharing method conspicuously improves the system performance up to around 20%–30% in SDWN, and the proposed algorithm achieves more efficient performance.
基金supported by the National Postdoctoral Science Foundation of China(2014M550068)
文摘Large latency of applications will bring revenue loss to cloud infrastructure providers in the cloud data center. The existing controllers of software-defined networking architecture can fetch and process traffic information in the network. Therefore, the controllers can only optimize the network latency of applications. However, the serving latency of applications is also an important factor in delivered user-experience for arrival requests. Unintelligent request routing will cause large serving latency if arrival requests are allocated to overloaded virtual machines. To deal with the request routing problem, this paper proposes the workload-aware software-defined networking controller architecture. Then, request routing algorithms are proposed to minimize the total round trip time for every type of request by considering the congestion in the network and the workload in virtual machines（VMs）. This paper finally provides the evaluation of the proposed algorithms in a simulated prototype. The simulation results show that the proposed methodology is efficient compared with the existing approaches.
文摘针对传统的IP欺骗攻击缓解方法存在运算开销大、缺乏灵活性等问题,提出了一种基于动态限制策略的软件定义网络(software defined network,SDN)中IP欺骗攻击缓解方法。首先,利用Packet-In消息中三元组信息回溯攻击路径,定位IP欺骗攻击源头主机;然后,由控制器制定动态限制策略对连接攻击源头主机的交换机端口的新流转发功能进行限制,待限制期满再恢复其转发新流的功能,限制期的大小随着被检测为攻击源的次数而增长。研究结果表明:这种动态的限制策略可阻隔攻击流进入SDN网络,从而有效避免SDN交换机、控制器以及链路过载;由于在限制期间无需再对这些限制的交换机端口进行实时监测,该方法在应对长时攻击时较传统方法具有更高的缓解效率和更少的资源消耗。
文摘针对工业物联网中业务需求多样性和服务质量(Quality of Service,QoS)要求差异性导致的网络资源利用低问题,提出一种基于深度强化学习的网络切片资源分配策略。该策略运用深度强化学习优化网络切片资源分配的准入控制,通过智能体在特定时间窗口内处理资源请求,并根据不同网络切片的QoS要求及请求准入结果进行资源的动态分配。实验结果表明,所提策略相比基准算法在提高网络收益、资源利用率和接收率方面分别提升了8.33%、9.84%和8.57%。该策略能够在保证服务质量的同时提高整个网络的效率和性能。
文摘软件定义网络(Software Defined Network,SDN)是一种新型的网络架构,相比于传统网络,简化了网络管理并更好地支持网络流量的动态控制,现已被许多应用领域采用。为了增强SDN的异常检测与防御能力,在拜占庭容错机制的基础上,提出一种异常检测方法,对异常或错误的指令容错,保证正确下发流表,同时通过理论分析证明该方法的有效性和安全性。实验表明,在SDN网络环境中,该检测方法能够快速检测出异常网络设备,降低SDN异常检测中的漏报率和误报率。
文摘针对SDN流量工程中流量预测基于静态时空依赖的问题,提出了一种基于注意力机制的图卷积神经网络(GCN)与门控递归单元(GRU)集成的动态网络流量预测方法——AGCNGRU(attention mechanism for GCNGRU model)。借助GCN捕获网络中节点之间的流量空间依赖性和GRU捕获流量经过网络中各节点的时间依赖性,通过时间注意力机制设计每个隐藏状态的权重,以调整时间点流量信息的重要性,同时通过数据驱动空间注意力机制动态自适应调整Laplace矩阵,实现动态提取网络信息数据时空相关性,最终完成动态流量精准预测。在GEANT的数据集上的实验表明,所提出的方法在均方误差方面比GCNGRU减少24.8%,比GRU减少66.4%,并通过与传统路由算法OSPF、DDPG算法比较,在90%的流量负载强度下,网络性能比OSPF提升了24%,比DDPG提升了8.1%,进一步说明了AGCNGRU算法网络流量准确预测带来的时效性和有效性。
文摘随着智能电网的快速发展,配电网中信息物理耦合关系日益紧密。这种耦合性使得配电网更容易被多方面极端事件所影响,在通信网络发生故障时会降低系统的态势感知和控制能力,从而制约配电网的灾后负荷恢复能力,因此通信网络恢复对灾后配电网负荷恢复至关重要。该文提出一种通信网络恢复和负荷恢复的协同优化决策方案,该方案将环网通信网络与软件定义网络(software defined networking,SDN)技术相结合,灵活恢复灾后的配电网通信网络,进而控制配电网拓扑重构形成以分布式电源为中心的微电网以恢复负荷电力供应,并进一步使用一种信息物理协同的启发式计算方法实现恢复方案的快速计算。最后,使用IEEE 33节点和IEEE 123节点测试系统验证所提出方法的优点和有效性。
