Research And Design Of Service Flow Classification Routing Strategy Of Internet Of Things Based On SDN

With the rapid development of Internet of things technology and the increasing number of Internet of things devices,the types of services in the network are becoming richer and richer,and a large amount of data traffic generated by various services also follows.The business traffic generated by different Internet of things devices has different requirements for the underlying resources of the network.The sharp increase in the types and quantity of business traffic has brought great challenges to the management of Internet of things and the guarantee of user service quality.Traditional networks only provide best effort services and use the number of hops between nodes as the standard to evaluate the advantages and disadvantages of scheduling paths,resulting in uneven resource allocation in the network,which is difficult to meet the quality of service requirements of Internet of things users for different services.As a new network architecture with the advantages of centralized control and transfer control separation,software defined network provides a new way to improve the management of the Internet of things and ensure the service quality of Internet of things users.This paper mainly studies and designs the service flow classification routing strategy of the Internet of things under the SDN architecture: the traffic sensing node is designed,and the network topology and resource information of the data plane are reported to the control plane in real time by using the characteristics of SDN transfer separation,so as to provide a basis for the classification and scheduling of service flow;Aiming at the characteristics of large amount of traffic data,many continuous characteristic attributes,unbalanced distribution of business categories and a large amount of noise in the Internet of things,a business flow classification model based on the combination of decision tree and integrated learning is designed and deployed in the control plane of SDN to classify the collected business flow information;According to the different requirements of different Internet of things services for various network resources,a service flow scheduling strategy based on network state Qo S parameters is designed.The correlation degree between services and network resources is calculated by deploying the strategy to the SDN control plane,and the correlation degree is used as the basis for Dijestra algorithm for routing scheduling,so as to select the optimal transmission path for service flow and realize routing scheduling by sending flow table,So as to ensure the service quality of Internet of things users.Finally,through experiments,the classification model proposed in this paper is compared with the other four classification models,and the classification accuracy,training time and classification time are significantly improved,which proves that the model has good classification performance;The service flow scheduling strategy designed in this paper is compared with the other two service flow scheduling strategies.The performance of this method is measured by the total delay,bandwidth and packet loss rate of the service flow from the source address to the destination address.The three indexes are significantly improved,which proves the superiority of this method.