Real-time Video Quality of Experience Monitoring for HTTPS and QUIC

Mazhar, M. Hammad; Shafiq, Zubair

doi:10.1109/infocom.2018.8486321

Cited by 82 publications

(71 citation statements)

References 21 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Recently, authors in [4], [9] use machine learning to predict QoE metrics of stalls, quality of playout and its variations using network traffic measurements such as RTT, bandwidth delay product, throughput and interarrival times. In another similar work [6] based on data collected in the lab, the authors use transport and network layer measurements to infer QoE impairments of startup delay, stalls and playback quality (three levels) in windows of 10 second duration. The prior works do not provide any estimation of the subjective MOS, rather they provide ML models for estimating the objective QoE metrics only.…”

Section: Related Workmentioning

confidence: 99%

“…In our approach, we play out a wide range of videos (considering a case of YouTube) under emulated network conditions to build a dataset that maps the enforced network QoS to QoE. Prior works [4], [6] have shown good performance of machine learning in the inference of application QoS features from encrypted traffic (e.g., stalls and startup delay). However they do not provide any subjective QoE prediction.…”

Section: Introductionmentioning

confidence: 99%

“…2) To the best of our knowledge, this is a first attempt at linking encrypted video traffic measurements to subjective MOS based on models such as ITU-T P.1203. Prior works [4], [6], [9] do not consider the subjective MOS, they consider objective QoE metrics only. 3) We provide a detailed performance comparison for ML modeling (classification and regression) with different types of network feature sets.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

From Network Traffic Measurements to QoE for Internet Video

Khokhar¹,

Ehlinger²,

Barakat³

2019

2019 IFIP Networking Conference (IFIP Networking)

View full text Add to dashboard Cite

Video streaming is a dominant contributor to the global Internet traffic. Consequently, monitoring video streaming Quality of Experience (QoE) is of paramount importance to network providers. Monitoring QoE of video is a challenge as most of the video traffic of today is encrypted. In this paper, we consider this challenge and present an approach based on controlled experimentation and machine learning to estimate QoE from encrypted video traces using network level measurements only. We consider a case of YouTube and play out a wide range of videos under realistic network conditions to build ML models (classification and regression) that predict the subjective MOS (Mean Opinion Score) based on the ITU P.1203 model along with the QoE metrics of startup delay, quality (spatial resolution) of playout and quality variations, and this is using only the underlying network Quality of Service (QoS) features. We comprehensively evaluate our approach with different sets of input network features and output QoE metrics. Overall, our classification models predict the QoE metrics and the ITU MOS with an accuracy of 63-90% while the regression models show low error; the ITU MOS (1-5) and the startup delay (in seconds) are predicted with a root mean square error of 0.33 and 2.66 respectively.

show abstract

Section: Related Workmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

From Network Traffic Measurements to QoE for Internet Video

Khokhar¹,

Ehlinger²,

Barakat³

2019

2019 IFIP Networking Conference (IFIP Networking)

View full text Add to dashboard Cite

show abstract

“…The wide adoption of end-to-end encryption has turned previous DPI-based approaches unreliable or even unfeasible, motivating a surge of papers focusing on the analysis of innetwork measurements through machine-learning models. For example, in [3], [13], authors apply different machine-learning approaches to estimate QoE-relevant metrics for YouTube by extracting features from the stream of encrypted packets, using simple features such as packet times and sizes, or throughput. Similarly, authors in [14] follow a machine-learning-based analysis to infer QoE metrics for YouTube streaming over cellular networks.…”

Section: Related Workmentioning

confidence: 99%

On the Analysis of YouTube QoE in Cellular Networks through in-Smartphone Measurements

Wassermann

Casas

Seufert

et al. 2019

2019 12th IFIP Wireless and Mobile Networking Conference (WMNC)

View full text Add to dashboard Cite

“…As this work is focused on continuous quality, the model only depends on the quality values of the last 15 seconds. [26] uses machine learning to predict initial delay, stalling, and video quality from the network traffic in windows of 10 s. The considered features are derived from IP or TCP/UDP headers only. ViCrypt [27] detects QoE degradations on encrypted video streaming traffic in real-time within 1 s by using a stream-like analysis approach with two continuous sliding windows and a cumulative window.…”

Section: Continuous Qualitymentioning

confidence: 99%

A Cumulative Quality Model for HTTP Adaptive Streaming

Tran

Ngoc

Hoβfeld

et al. 2018

2018 Tenth International Conference on Quality of Multimedia Experience (QoMEX)

View full text Add to dashboard Cite

Thanks to the abundance of Web platforms and broadband connections, HTTP Adaptive Streaming has become the de facto choice for multimedia delivery nowadays. However, the visual quality of adaptive video streaming may fluctuate strongly during a session due to bandwidth fluctuations. So, it is important to evaluate the quality of a streaming session over time. In this paper, we propose a model to estimate the cumulative quality for HTTP Adaptive Streaming. In the model, a sliding window of video segments is employed as the basic building block. Through statistical analysis using a subjective dataset, we identify three important components of the cumulative quality model, namely the minimum window quality, the last window quality, and the average window quality. Experiment results show that the proposed model achieves high prediction performance and outperforms related quality models. In addition, another advantage of the proposed model is its simplicity and effectiveness for deployment in real-time estimation. The source code of the proposed model has been made available to the public at https

show abstract

Real-time Video Quality of Experience Monitoring for HTTPS and QUIC

Cited by 82 publications

References 21 publications

From Network Traffic Measurements to QoE for Internet Video

From Network Traffic Measurements to QoE for Internet Video

On the Analysis of YouTube QoE in Cellular Networks through in-Smartphone Measurements

A Cumulative Quality Model for HTTP Adaptive Streaming

Contact Info

Product

Resources

About