Name
Abstract | ||
|---|---|---|
Network traffic classification is important for the management of network resource and the support quality of multimedia services. To realize the fine-grained classification of typical Internet video traffic, this paper studies and analyses the characteristics of video flow change in transmission process and the statistic characteristics of its main protocol data. According to different service models from the network services and the users’ demand of video quality, we propose two new sets of features for video traffic classification, including: downlink rate probability distribution model and main protocol data statistics. The experimental results show that the two sets of features can improve the performance of classification compared to existing methods. |
Year | DOI | Venue |
|---|---|---|
2019 | 10.1007/s11042-018-6965-6 | Multimedia Tools and Applications |
Keywords | Field | DocType |
Video feature, Traffic classification, Probability distribution, Feature mining | Traffic classification,Data mining,Statistic,Internet video,Pattern recognition,Computer science,Feature mining,Protocol data unit,Probability distribution,Artificial intelligence,Video quality,Telecommunications link | Journal |
Volume | Issue | ISSN |
78.0 | 12 | 1573-7721 |
Citations | PageRank | References |
0 | 0.34 | 17 |
Authors | ||
4 | ||
Name | Order | Citations | PageRank |
|---|---|---|---|
| Lingyun Yang | 1 | 0 | 1.01 |
| Yu-ning Dong | 2 | 17 | 3.05 |
| Wei Tian | 3 | 28 | 10.31 |
| Zaijian Wang | 4 | 10 | 3.70 |