Font Size: a A A

Research And Optimization Of Na Med Data Networks In Live Stre Aming Scenarios

Posted on:2020-11-28Degree:MasterType:Thesis
Country:ChinaCandidate:X S PengFull Text:PDF
GTID:2428330623456367Subject:Software engineering
Abstract/Summary:PDF Full Text Request
In recent years,due to the rapid development of the Internet,the information that people can obtain is more diversified.As a more vivid information transmission medium than text and pictures,audio and video have gradually become the indispensable means of information acquisition and communication in people's daily life and entertainment.With the decline of broadband tariffs and the development of CDN technology,real-time video services such as live streaming have begun to be accepted.Compared with the resources of ordinary editing and finishing video,the live streaming of the video is not dense,but the high real-time characteristics usually make users more engaged in the viewing process.Due to the high real-time performance and the large number of participants,the live video has higher requirements for the network environment.Since TCP/IP is an end-to-end communication method,the efficiency of video distribution has a great influence on the quality of live streaming services,so a specific server is needed to provide real-time video transmission for a large number of viewers at the same time.The complexity of the network environment in the live streaming scenario and the high concurrent requests of users are less efficient in the TCP/IP system for end-to-end transmission.Named Data Network(NDN)is a new type of network architecture that is developing rapidly.NDN is based on content-centric thinking,so it is very suitable for the needs of a large amount of information transmission in today's networks,so that people do not care about the specific location of the information obtained,but Concerned about the data content itself.NDN is very suitable for the distribution of large amounts of data.Live streaming is a scenario that needs to simultaneously share current resources to a large number of viewers.It is very suitable for implementation in NDN.The main research direction of this paper is to combine the NDN network with the live streaming scenario to explore the part that can be optimized in the live streaming scenario under NDN.Although from the perspective of content distribution,the NDN network can provide better data transmission services for the live streaming system,from the perspective of details,the way of pulling data packets in the NDN network is contrary to the transmission method of pushing live data in the live streaming scenario.Therefore,there will be some problems in the column;in addition,although the live data can be distributed to multiple consumers perfectly through the NDN,after analyzing the effects of PIT and CS,it can be seen that there is still room for optimization.This paper studies the live streaming scenario in the NDN network.Based on the live streaming system architecture under the current IP system,and referring to the role of each part,the architecture of the live streaming system in the NDN is designed and proposed according to the characteristics of the NDN network.The synchronization policy of the interest packet and the frame group forwarding policy are used to optimize the interaction mode of the interest packet and the data packet in the live streaming scenario.In terms of caching strategy,a bubbling caching strategy is proposed.According to the characteristics of the live streaming scenario in the NDN,a small portion of the data packet is cached in the network to determine the location of the overall data to be cached.The caching strategy proposed in this paper is applicable to scenarios where the real-time performance is strong and the data transmission topology is relatively stable.According to the simulation results of ndn SIM,the caching strategy proposed in this paper can effectively reduce the redundant cache in the network and effectively reduce redundant traffic in the live streaming scenario.
Keywords/Search Tags:NDN, Live broadcast, Cache policy, Forwarding strategy
PDF Full Text Request
Related items