ABSTRACT
Accurate cross-layer information is very useful to optimize mobile networks for specific applications. However, providing application-layer information to lower protocol layers has become very difficult due to the wide adoption of end-to-end encryption and due to the absence of cross-layer signaling standards. As an alternative, this paper presents a traffic profiling solution to passively estimate parameters of HTTP Adaptive Streaming (HAS) applications at the lower layers. By observing IP packet arrivals, our machine learning system identifies video flows and detects the state of an HAS client's play-back buffer in real time. Our experiments with YouTube's mobile client show that Random Forests achieve very high accuracy even with a strong variation of link quality. Since this high performance is achieved at IP level with a small, generic feature set, our approach requires no Deep Packet Inspection (DPI), comes at low complexity, and does not interfere with end-to-end encryption. Traffic profiling is, thus, a powerful new tool for monitoring and managing even encrypted HAS traffic in mobile networks.
- A. Molavi Kakhki et al. 2016. BingeOn Under the Microscope: Understanding T-Mobile's Zero-Rating Implementation. In Proc. ACM Internet-QoE. 43--48. Google ScholarCross Ref
- L. Breiman. 2001. Random Forests. Machine Learning 45, 1 (Oct. 2001), 5--32. Google ScholarDigital Library
- Cisco. 2017. Visual Networking Index: Global mobile data traffic forecast update, 2016-2021. White Paper (Feb. 2017).Google Scholar
- DASH Industry Forum. 2014. Guidelines for Implementation: DASH-AVC/264 Test cases and Vectors. Report (Jan. 2014).Google Scholar
- G. Dimopoulos, P. Barlet-Ros, and J. Sanjuàs-Cuxart. 2013. Analysis of YouTube user experience from passive measurements. In Proc. IEEE CNSM. 260--267.Google Scholar
- G. Dimopoulos, I. Leontiadis, P. Barlet-Ros, and K. Papagiannaki. 2016. Measuring Video QoE from Encrypted Traffic. In Proc. ACM IMC. 513--526. Google ScholarDigital Library
- Zweites Deutsches Fernsehen. 2017. Heute Show. Video. Retrieved May 21, 2018 from https://www.youtube.com/watch?v=N2sCbtodGMIGoogle Scholar
- Blender Foundation. 2013. Tears of steel. Video. Retrieved May 21, 2018 from https://www.youtube.com/watch?v=OHOpb2fS-cMGoogle Scholar
- C. Fraleigh et al. 2003. Packet-level traffic measurements from the Sprint IP backbone. IEEE Network 17, 6 (Nov. 2003), 6--16. Google ScholarDigital Library
- Y. Freund and R. E. Schapire. 1997. A Decision-Theoretic Generalization of On-Line Learning and an Application to Boosting. J. Comput. System Sci. 55, 1 (Aug. 1997), 119--139. Google ScholarDigital Library
- S. Galetto et al. 2017. Detection of video/audio streaming packet flows for non-intrusive QoS/QoE monitoring. In Proc. IEEE MN. 1--6.Google ScholarCross Ref
- ISO/1EC. 2014. Dynamic adaptive streaming over HTTP (DASH). International Standard 23009-1:2014 (May 2014).Google Scholar
- T. Karagkioules, C. Concolato, D. Tsilimantos, and S. Valentin. 2017. A Comparative Case Study of HTTP Adaptive Streaming Algorithms in Mobile Networks. In Proc. ACM NOSSDAV. 1--6. Google ScholarDigital Library
- M. Katsarakis, R. C. Teixeira, M. Papadopouli, and V. Christophides. 2016. Towards a Causal Analysis of Video QoE from Network and Application QoS. In Proc. ACM Internet-QoE. 31--36. Google ScholarCross Ref
- V. Krishnamoorthi, N. Carlsson, E. Halepovic, and E. Petajan. 2017. BUFFEST: Predicting Buffer Conditions and Real-time Requirements of HTTP(S) Adaptive Streaming Clients. In Proc. ACM MMSys. 76--87. Google ScholarDigital Library
- Y. Lin and Y. Jeon. 2006. Random Forests and Adaptive Nearest Neighbors. J. Amer. Statist. Assoc. 101, 474 (2006), 578--590.Google ScholarCross Ref
- A. Mondal et al. 2017. Candid with YouTube: Adaptive Streaming Behavior and Implications on Data Consumption. In Proc. ACM NOSSDAV. 19--24 Google ScholarDigital Library
- T. T. T. Nguyen and G. Armitage. 2008. A survey of techniques for Internet traffic classification using machine learning. IEEE Communications Surveys Tutorials 10, 4 (Fourth 2008), 56--76. Google ScholarDigital Library
- National Institute of Fundamental Studies Sri Lanka. 2016. Science Copies Nature's Secrets. Video. Retrieved May 21, 2018 from https://www.youtube.com/watch?v=2d1VrCvdzbYGoogle Scholar
- I. Orsolic, D. Pevec, M. Suznjevic, and L. Skorin-Kapov. 2017. A machine learning approach to classifying YouTube QoE based on encrypted network traffic. Springer Multimedia Tools and Applications 76, 21 (Nov. 2017), 22267--22301. Google ScholarDigital Library
- R. Pantos and W. May. 2011. HTTP live streaming. IETF, Informational Internet-Draft 2582 (Sept. 2011).Google Scholar
- J. Roskind. 2013. QUIC: Multiplexed Stream Transport over UDP. Design Document and Specification Rationale (Dec. 2013).Google Scholar
- J. Samain et al. 2017. Dynamic Adaptive Video Streaming: Towards a Systematic Comparison of ICN and TCP/IP. IEEE Transactions on Multimedia 19, 10 (Oct. 2017), 2166--2181.Google ScholarCross Ref
- Sandvine. 2016. Global Internet Phenomena: Latin America & North America. Report (June 2016).Google Scholar
- C. Seiffert, T. M. Khoshgoftaar, J. Van Hulse, and A. Napolitano. 2010. RUSBoost: A Hybrid Approach to Alleviating Class Imbalance. IEEE Transactions on Systems, Man, and Cybernetics - Part A: Systems and Humans 40, 1 (Jan. 2010), 185--197. Google ScholarDigital Library
- tc. 2001. traffic configuration tool (tc). Retrieved May 21, 2018 from https://lartc.org/manpages/tc.htmlGoogle Scholar
- Tcpdump. 2015. tcpdump packet analyzer v. 4.7.4, (libpeap v. 1.7.4). Retrieved May 21, 2018 from https://www.tcpdump.orgGoogle Scholar
- The Math Works Inc. 2015. Statistics and Machine Learning Toolbox v10.1. Retrieved May 21, 2018 from https://mathworks.com/products/statistics.htmlGoogle Scholar
- D. Tsilimantos, T. Karagkioules, A. Nogales-Gómez, and S. Valentin. 2017. Traffic profiling for mobile video streaming. In Proc. IEEE ICC. 1--7.Google Scholar
- D. Tsilimantos, A. Nogales-Gómez, and S. Valentin. 2016. Anticipatory Radio Resource Management for Mobile Video Streaming with Linear Programming. In Proc. IEEE ICC. 1--6.Google Scholar
- T-Mobile USA. 2016. Content Provider Technical Requirements for Binge On. (March 2016). Retrieved May 21, 2018 from https://www.t-mobile.com/content/dam/tmo/en-g/pdf/BingeOn-Video-Technical-Criteria-March-2016.pdfGoogle Scholar
- D. De Vleeschauwer et al. 2013. Optimization of HTTP adaptive streaming over mobile cellular networks. In Proc. IEEE INFOCOM. 898--997.Google Scholar
- YouTube. 2015. YouTube RED. Retrieved May 21, 2018 from https://www.youtube.com/redGoogle Scholar
Index Terms
- Classifying flows and buffer state for youtube's HTTP adaptive streaming service in mobile networks
Recommendations
Performance of Low-Latency HTTP-based Streaming Players
MHV '23: Proceedings of the 2nd Mile-High Video ConferenceReducing end-to-end streaming latency is critical to HTTP-based live video streaming. There are currently two main technologies in this domain: Low-Latency HTTP Live Streaming (LL-HLS) and Low-Latency Dynamic Adaptive Streaming over HTTP (LL-DASH). ...
Performance of Low-Latency HTTP-based Streaming Players
MMSys '21: Proceedings of the 12th ACM Multimedia Systems ConferenceReducing end-to-end streaming latency is critical to HTTP-based live video streaming. There are currently two technologies in this domain: Low-Latency HTTP Live Streaming (LL-HLS) and Low-Latency Dynamic Adaptive Streaming over HTTP (LL-DASH). Many ...
A Comparative Case Study of HTTP Adaptive Streaming Algorithms in Mobile Networks
NOSSDAV'17: Proceedings of the 27th Workshop on Network and Operating Systems Support for Digital Audio and VideoHTTP Adaptive Streaming (HAS) techniques are now the dominant solution for video delivery in mobile networks. Over the past few years, several HAS algorithms have been introduced in order to improve user quality-of-experience (QoE) by bit-rate ...
Comments