【摘要】 物联网环境下LTE(long term evolution)网络的上行流量为主要流量,网络服务的类型较多。提出了一种自适应的流量公平调度算法,控制了物联网大数据流导致LTE网络拥塞的问题。在时间域数据包的调度中,设计了一种M2M(machine to machine)设备数量的控制机制,避免了M2M通信对H2H通信的影响,调度器使用资源分配的历史记录来估计当前H2H设备的资源需求,并为H2H设备保留充足的资源。在频率域数据包调度中,根据拥塞的级别动态地选择M2M设备的最大数量,从而缓解网络拥塞导致事件驱动型设备的延迟现象。仿真实验结果表明:本调度器保证了网络流量的QoS,实现了较高的公平性,从而缓解了"饿死"现象的发生。更多还原

【Abstract】 The uplink traffic becomes the main traffic and there are many types of network services in LTE network in the scenario of Internet of Things,an adaptive traffic fair scheduler for LTE network is proposed to control the congestion of LTE network because of introducing the Internet of Things. In the time domain packet scheduling process,a mechanism for controlling the count of M2 M devices is designed to prevent the influence to the H2 H communication by M2 M communication. The historic information of resource allocation is used by the scheduler to estimate the current resource requirement of the H2 H devices,and enough resource are reserved for H2 H devices. In the frequency domain packet scheduling process,the maximum count of M2 M devices is selected dynamically according to the congestion levels of the network to reduce the delay of event-driven devices caused by network congestion. Simulation experimental results show that the proposed scheduler satisfies the QoS requirements of LTE network,and realizes a good fairness to avoid the problem of starvation.更多还原