Research On Real Time Service Flow Control Method For Information Centric Networking

Information-Centric Networking(ICN)takes information as the center,decoupling data and location,so that network nodes are tightly coupled,transforming the traditional end-to-end communication mode into efficient content distribution and acquisition,which is expected to fundamentally solve the efficient access of users to information.In real-time services,requests are time-sensitive,and how to effectively dispatch traffic and improve network performance has become a key issue in ICN networks.Given the above problems,we conduct in-depth research from the perspective of service credibility.For real-time services,the traffic scheduling problem is studied progressively,and the main work and contributions of this paper are as follows.1.In the ICN cable network,a Long-term Deadline Fair Oblivious(LDFO)strategy is proposed for the user’s delay tolerance for different services in real-time services.Through the combination of flow control and the Lyapunov drift method,an appropriate queue is established to complete user requests and data returns.By calculating the queue backlog difference for each node,the proposed LDFO strategy can assign a transmission rate to each request to maximize throughput.According to the user’s delay tolerance for different requests and the freshness of information,different services are given priority and network resources are allocated.All queue backlogs are bounded and can provide latency guarantees,so the algorithm can achieve good performance in a stable queue system.Experiments show that the algorithm has a better performance guarantee while ensuring lower complexity.2.In the ICN wired network,considering the popularity of different contents in real-time services and considering the issue of flow fairness,a Routing Load Forwarding(RLF)strategy is proposed.This strategy utilizes the popularity of different contents,firstly packages the data into a stream,and at the same time uses the Lyapunov drift method to construct a queuing system,through which the request forwarding and return data are jointly completed.According to the popularity of different content,a cache priority strategy is constructed,and the data forwarding rate is adjusted according to the routing load and through the cooperation between nodes.At the same time,active PUSH is used to negotiate with the responder,to adjust the source end,balance the network load,and improve the network performance.Experiments show that the algorithm has a good performance guarantee.