节点文献
支持QoS的CPS网络路由算法的研究与实现
Research and Implementation of QoS Supported Routing Algorithm for CPS Network
【作者】 李雪梅;
【导师】 卢国明;
【作者基本信息】 电子科技大学 , 工程硕士(专业学位), 2017, 硕士
【摘要】 信息物理融合系统(Cyber-physical systems,CPS)是由感知、计算和控制等多个子系统组成的复杂网络系统,系统中包含感知数据、计算结果、控制指令等多种类型的数据。CPS通常应用于野外环境监测、基础设施控制、智慧交通等大型工程系统,其数据规模较物联网等系统更大。因此,数据传输的实时性和可靠性成为影响CPS系统性能的关键因素之一。同时,CPS的通信范围覆盖系统的物理层、网络层和决策层,各层的网络类型不一样,通信需求也不一样,因此,一个合理的网络结构和满足相应QoS需求的路由算法对CPS系统尤为重要。针对以上问题,本文提出了相关解决方案,主要工作及创新点如下:(1)提出一个基于跨层QoS模型的数据调度机制。其主要目的是通过节点数据选择来满足CPS系统实时业务数据的传输需求。该算法的核心思想是提出一个跨层QoS模型,分别对节点、网络和业务数据的QoS需求进行分析,将各层的QoS需求进行量化,然后以满足这些QoS需求为目标设计一个数据选择控制策略。当源节点需要传输数据时,应用该策略来选择数据的发送序列,这样节点就可以通过对网络资源的合理利用来满足业务数据的QoS需求,同时提高网络传输的价值和网络资源分配的公平性。仿真实验表明,使用该数据调度机制的路由算法比不使用该数据调度机制的路由算法在网络效用和网络公平性上都有明显提升。(2)提出一个适用于CPS网络的QoS路由算法,该算法主要目的是解决网络传输的实时性和可靠性问题。算法的核心思想是选取网络传输中的延迟和可靠性两个QoS指标,将其分别量化为一个阈值。以机会路由为基础,当源节点需要传输数据时,首先以满足延迟指标为目的,对其所有邻居节点区分优先级排序;接着以满足可靠性指标为目的,在邻居节点的优先级列表中根据策略选择部分节点作为候选节点集;集合中节点之间通过协同,确定相应中继节点来转发数据。仿真实验表明,该路由算法在延迟和投递率的性能较同类型的QORM路由算法更优。(3)对CPS系统的网络结构进行了合理的设计,构建了一个以容迟/容断网络(delay/disruption tolerant network,DTN)作为组网方式的异构网络结构。并在ONE仿真平台上设计实现了以该结构为基础的CPS的系统模型,将上述两个算法在该平台上进行了验证实现。
【Abstract】 Cyber-physical system is a complex network system which is composed by multiple subsystems,such as sensing,computing and control system.The system contains various types of data such as perceptual data,calculation results and control instructions.CPS is usually used in wild environmental monitoring,infrastructure control,intelligent transportation and other large engineering systems,the amount of data in CPS is greater than the Internet of things.Therefore,the delay and reliability of data transmission become one of the key factors that affect the performance of CPS.At the same time,the communication range of CPS covers the physical layer,network layer and decision layer of the system.The network type of each layer is different and the communication demand is different too.Therefore,a complete and reasonable network structure and routing algorithm which is satisfying the corresponding QoS requirements are particularly important to CPS.The main work and innovation as follows:(1)A data scheduling mechanism based on cross-layer QoS model is proposed.Its main purpose is to meet the transmission requirements of real-time service data of CPS system through node’s data scheduling.The core idea of this algorithm is to propose a cross-layer QoS model to analyze the QoS requirements of nodes,networks and services data respectively.The QoS requirements of each layer are quantified by mathematical formula,and then a data scheduling mechanism is designed to meet these QoS requirements.When the source node needs to transmit data,this strategy is used to determine the transmission sequence of the data,so that the node can meet the QoS requirement of the service data through the rational use of the network resources,and improve the value of the network transmission and the fairness of the network resource allocation.The simulation results show that the routing algorithm using the data scheduling mechanism has improved the network utility and the network fairness than the routing algorithm which does not use the data scheduling mechanism.(2)A QoS routing algorithm for CPS network is proposed.The main purpose of this algorithm is to solve the delay and reliability of network transmission.The core idea of the algorithm is to select the delay and reliability as two QoS indicators of network transmission,and quantify them into two thresholds.Based on the opportunistic routing,when the source node needs to transmit data,the maximum possible transmission delay of all neighbor nodes to the destination node is calculated for the purpose of satisfying the delay,and the neighbor nodes are sorted according to the delay.And then select the node as the candidate node set according to the policy in the priority list of the neighbor node for the purpose of satisfying the reliability index.The nodes in the collection cooperate to determine the corresponding node to forward the data.The simulation results show that the proposed algorithm has better performance in the delay and delivery success rate than the QORM routing algorithm with the same type.(3)Finally,the network structure of CPS system is designed reasonably,and a heterogeneous network structure with DTN as networking is constructed.The system model of CPS based on this structure is designed and implemented on ONE simulation platform,and the above two algorithms are verified on the platform.
【Key words】 CPS network; QoS; data scheduling; opportunistic routing; ONE simulator;
- 【网络出版投稿人】 电子科技大学 【网络出版年期】2018年 02期
- 【分类号】TN915.0
- 【下载频次】76