Research On Congestion Control Algorithm Based On Cache Strategy In Named Data Networking

Since the 1960s,the Internet based on the TCP/IP protocol has played an increasingly important role in modern society.The TCP/IP network architecture is a host-centric model that was developed based on the early Internet application model,Such as providing connectivity and sharing resources.However,with the rapid development of computer technology and network applications,the network transmission mode has also changed from resource sharing to content distribution and acquisition.The focus of people's attention has changed from "where to obtain content" to "what content to obtain".However,the TCP/IP network originally designed for end-to-end communication is difficult to adapt to this change.The limitations of the TCP/IP network limit the Internet to provide users with higher-performance services.Therefore,in order to improve network performance and provide better service quality,Named Data Networking(NDN)proposes to build a new network architecture with content as the center,and realize content retrieval and data transmission by naming information,while Named Data Networking's characteristics of connectionless and network caching greatly improve network performance.Proposing a congestion control algorithm for NDN is one of the important research directions of NDN,and when considering AR/VR and other large data delay sensitive applications,NDN's One-Interest-One-Data transmission mode cannot achieve the best transmission efficiency,so this paper designs a congestion control mechanism controlled by the content provider based on the One-Interest-Multiple-Data transmission mode.In addition,node caching(CS)is an important feature of NDN.Node caching can increase the speed of resource acquisition,but it will also adversely affect the accuracy of the congestion control mechanism.Therefore,before designing the congestion control algorithm,this article first proposes a probabilistic caching strategy based on node importance,content popularity,fragment cache value,and a cache update strategy based on content fragment retention value and content fragment redundancy,which reduces content fragments Redundancy,and aggregating segments belonging to the same content in a few nodes,which facilitates the design of congestion control algorithms.The main work of this paper is as follows:1.This paper proposes a probabilistic cache strategy and corresponding cache update strategy.When the node receives the content fragment,it calculates the caching probability based on the cache value of the content fragment,the popularity of the corresponding content,and the node importance.Each content segment in the cache establishes two attributes of segment retention value and redundancy to measure its life cycle in the cache.When the space occupied by the cached content exceeds the rated threshold,the cache update strategy is executed based on the segment retention value and redundancy,thereby reducing the redundancy of the cached content and improving the cache efficiency.2.This paper proposes a congestion control algorithm based on bandwidth and delay awareness.The algorithm uses interest packets to collect the bottleneck link bandwidth and delay information on the path from the content requester to the content provider,and the content provider calculates the congestion window and sending rate based on this information.In view of the node caching characteristics of NDN,a cache marking method is proposed.When the intermediate node has a partial cache of the requested content,the cache content of the node is marked,so that the cached content of each node can be transmitted in an orderly manner,making full use of the node cache and reducing time required for transmission will not affect the accuracy of the congestion control algorithm.3.This paper verifies the performance of the caching strategy and congestion control algorithm under the ndn SIM simulation platform,and selects several classic and mainstream algorithm strategies for comparison accordingly.Experimental results show that the proposed caching strategy has better performance in terms of cache hit rate and average hit hops.At the same time,it can achieve lower cache content fragment redundancy and higher network cache quantity.The proposed congestion control algorithm can achieve better transmission performance,more stable transmission rate,can reduce the time required for transmission,and will not be affected by the node cache to the accuracy of the congestion control algorithm,and can use the node cache to increase the transmission rate.The congestion control algorithm running on the basis of the proposed cache strategy can achieve higher transmission efficiency and reduce transmission overhead.