Uplink vs. Downlink: Machine Learning-Based Quality Prediction for HTTP Adaptive Video Streaming
Streaming video is responsible for the bulk of Internet traffic these days. For this reason, Internet providers and network operators try to make predictions and assessments about the streaming quality for an end user. Current monitoring solutions are based on a variety of different machine learning...
Main Authors: | , , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
MDPI AG
2021-06-01
|
Series: | Sensors |
Subjects: | |
Online Access: | https://www.mdpi.com/1424-8220/21/12/4172 |
id |
doaj-c18a1ef9fa1c4105ba9e76d74f4ccdb4 |
---|---|
record_format |
Article |
spelling |
doaj-c18a1ef9fa1c4105ba9e76d74f4ccdb42021-07-01T00:28:33ZengMDPI AGSensors1424-82202021-06-01214172417210.3390/s21124172Uplink vs. Downlink: Machine Learning-Based Quality Prediction for HTTP Adaptive Video StreamingFrank Loh0Fabian Poignée1Florian Wamser2Ferdinand Leidinger3Tobias Hoßfeld4Institute of Computer Science, University of Würzburg, 97074 Würzburg, GermanyInstitute of Computer Science, University of Würzburg, 97074 Würzburg, GermanyInstitute of Computer Science, University of Würzburg, 97074 Würzburg, GermanyInstitute of Computer Science, University of Würzburg, 97074 Würzburg, GermanyInstitute of Computer Science, University of Würzburg, 97074 Würzburg, GermanyStreaming video is responsible for the bulk of Internet traffic these days. For this reason, Internet providers and network operators try to make predictions and assessments about the streaming quality for an end user. Current monitoring solutions are based on a variety of different machine learning approaches. The challenge for providers and operators nowadays is that existing approaches require large amounts of data. In this work, the most relevant quality of experience metrics, i.e., the initial playback delay, the video streaming quality, video quality changes, and video rebuffering events, are examined using a voluminous data set of more than 13,000 YouTube video streaming runs that were collected with the native YouTube mobile app. Three Machine Learning models are developed and compared to estimate playback behavior based on uplink request information. The main focus has been on developing a lightweight approach using as few features and as little data as possible, while maintaining state-of-the-art performance.https://www.mdpi.com/1424-8220/21/12/4172HTTP adaptive video streamingquality of experience predictionmachine learning |
collection |
DOAJ |
language |
English |
format |
Article |
sources |
DOAJ |
author |
Frank Loh Fabian Poignée Florian Wamser Ferdinand Leidinger Tobias Hoßfeld |
spellingShingle |
Frank Loh Fabian Poignée Florian Wamser Ferdinand Leidinger Tobias Hoßfeld Uplink vs. Downlink: Machine Learning-Based Quality Prediction for HTTP Adaptive Video Streaming Sensors HTTP adaptive video streaming quality of experience prediction machine learning |
author_facet |
Frank Loh Fabian Poignée Florian Wamser Ferdinand Leidinger Tobias Hoßfeld |
author_sort |
Frank Loh |
title |
Uplink vs. Downlink: Machine Learning-Based Quality Prediction for HTTP Adaptive Video Streaming |
title_short |
Uplink vs. Downlink: Machine Learning-Based Quality Prediction for HTTP Adaptive Video Streaming |
title_full |
Uplink vs. Downlink: Machine Learning-Based Quality Prediction for HTTP Adaptive Video Streaming |
title_fullStr |
Uplink vs. Downlink: Machine Learning-Based Quality Prediction for HTTP Adaptive Video Streaming |
title_full_unstemmed |
Uplink vs. Downlink: Machine Learning-Based Quality Prediction for HTTP Adaptive Video Streaming |
title_sort |
uplink vs. downlink: machine learning-based quality prediction for http adaptive video streaming |
publisher |
MDPI AG |
series |
Sensors |
issn |
1424-8220 |
publishDate |
2021-06-01 |
description |
Streaming video is responsible for the bulk of Internet traffic these days. For this reason, Internet providers and network operators try to make predictions and assessments about the streaming quality for an end user. Current monitoring solutions are based on a variety of different machine learning approaches. The challenge for providers and operators nowadays is that existing approaches require large amounts of data. In this work, the most relevant quality of experience metrics, i.e., the initial playback delay, the video streaming quality, video quality changes, and video rebuffering events, are examined using a voluminous data set of more than 13,000 YouTube video streaming runs that were collected with the native YouTube mobile app. Three Machine Learning models are developed and compared to estimate playback behavior based on uplink request information. The main focus has been on developing a lightweight approach using as few features and as little data as possible, while maintaining state-of-the-art performance. |
topic |
HTTP adaptive video streaming quality of experience prediction machine learning |
url |
https://www.mdpi.com/1424-8220/21/12/4172 |
work_keys_str_mv |
AT frankloh uplinkvsdownlinkmachinelearningbasedqualitypredictionforhttpadaptivevideostreaming AT fabianpoignee uplinkvsdownlinkmachinelearningbasedqualitypredictionforhttpadaptivevideostreaming AT florianwamser uplinkvsdownlinkmachinelearningbasedqualitypredictionforhttpadaptivevideostreaming AT ferdinandleidinger uplinkvsdownlinkmachinelearningbasedqualitypredictionforhttpadaptivevideostreaming AT tobiashoßfeld uplinkvsdownlinkmachinelearningbasedqualitypredictionforhttpadaptivevideostreaming |
_version_ |
1721348537537527808 |