Research On Edge Caching Technology For Highly Scalable Encoded Video Transmission

With the rapid growth of related Internet services such as self-media and short videos,videorelated service traffic accounts for an increasing proportion of network traffic.In the current process of video services,video storage and transmission capacity bottlenecks are gradually becoming a key issue to limit the quality of streaming media services,so research on video storage and transmission has a very positive impact on improving the quality of video services.Video coding technology is to reduce the overhead in the video transmission process and compress the original video content of a technology,there are a number of different video coding methods to compress the transmission of video,the principle and characteristics of different coding methods are also different.In this paper,two representative coding methods of advanced video coding AVC and scalability video coding SVC are selected as objects,and on the basis of in-depth analysis of the coding principles of the two coding methods,several comparative experiments are carried out from three different performance angles of compression bit rate,fault tolerance under different network transmission conditions and space occupation of cache video,and according to the experimental results,the advantages and disadvantages of the two coding methods in different performance fields,as well as some problems existing in different coding methods.In order to shorten the latency generated by users when requesting content from the server,edge caching technology has been rapidly developed in recent years,which reduces the time consumption generated during content transmission by caching the content requested by users to edge devices such as base stations close to users.However,due to the limited capacity of the edge device,the selection of the appropriate cache content in the edge cache process has an important impact on the hit ratio of the edge cache.This paper designs a cache selection algorithm SVCBEC that uses SVC for video encoding,which predicts the future popular videos to be cached according to the past popularity of the requested content,and then proposes a series of edge cache service evaluation parameters to quantify the actual user experience of the edge caching scheme.Finally,the measured data is used to simulate the algorithm,and the experimental results using the SVC encoding method are compared with the results using the AVC encoding method,and the results show that the SVC encoding method has excellent performance in this environment.High-scalability encoding fine-grained SVC is a new variant of scalability coding that has emerged in recent years,which can freely adjust the number of layers and bitrate allocation of the cache,and has stronger adaptability and practicality for complex situations.In order to apply this encoding technology to the environment of edge caching,this paper designs a cache selection algorithm EC-FGSVC with high scalability encoding,which improves the flexibility of caching,so that the edge device can freely adjust the size of the cache video when caching,and compares the algorithm results with the commonly used cache replacement algorithms such as LRU,LFU and FIFO through experiments,and the results show that the algorithm has a great advantage over the algorithm in terms of device service score.Multiple edge caches in a region can form an edge cache system,when the user’s request misses one of the devices,the device can pull the content required by the user from other caches in the system,which can effectively reduce the additional delay caused by requesting content to the CDN server and link resource consumption.Based on this idea,this paper proposes a cache allocation algorithm CSB-VP-HSE suitable for edge caching systems.The algorithm reduces the number of duplicate caches and improves the diversity of cache content in the system by introducing the interaction between devices in a single edge system,and the experimental results show that the algorithm effectively improves the overall video service experience of users in the system while ensuring the hit rate.