Research On Intelligent Context Awareness Technology And Routing Optimization Application In SDN

With the development and breakthrough of network technology,the network services become increasingly complex and diverse.If the real-time detection of the running state in the network cannot be controlled,the management of the network will undoubtedly face enormous challenges.Nevertheless,the traditional Internet architecture resource management mode is fixed.It not only cannot meet the high demand of current network services,but cannot meet the real-time state awareness requirements of network managers as well.Software Defined Network and SiFiNet provides a powerful solution to the current network dilemma.The SiFiNet was proposed for a short period of time and it remains to be studied.Based on the software-defined network architecture,this thesis studies the various scenario elements in the software-defined network,including the basic environment,resources,and user services of the network.The formal description method is used to analyze various scenario elements,and a unified description method is established,including the basic scenario description method,the resource scenario description method,and the business scenario description method.On the basis of the established description method,this thesis proposes an intelligent context awareness scheme based on deep learning,defines the expression form of the perception results,and proposes an intelligent context awareness framework in the SDN environment.Based on the proposed framework,this dissertation focuses on the SDN intelligent context awareness model.In order to achieve context awareness in the SiFiNet,a self-encoder is integrated into RouteNet to propose an SDN intelligent context awareness model based on graph neural networks.Experiments prove that the proposed model can effectively reduce the perception error.By perceiving the QoS performance and resource distribution in the network,and combining deep reinforcement learning technology,a routing optimization mechanism based on GGAE model and DDPG algorithm is proposed in this thesis.That can be used to verify the performance of the proposed SDN intelligent context awareness model in network optimization.Experimental results show that the SDN intelligent context awareness model has a substantial effect on the network optimization and achieves the goal of network optimization.