Rebera: Real-time Bandwidth Estimation and Rate Adaptation for Video Calls over Cellular Networks

HomePage.Rebera History

Show minor edits - Show changes to output

July 20, 2016, at 12:58 AM EST by 184.152.50.28 -
Changed line 29 from:
In our emulations, we used [[ https://github.com/keithw/multisend/archive/evaluated.zip | Cellsim ]], an open-source variable-capacity link emulator. You '''really''' should check out their main project [[http://alfalfa.mit.edu | Sprout]] as well. We wanted to be able to compare Rebera
to:
In our emulations, we use [[ https://github.com/keithw/multisend/archive/evaluated.zip | Cellsim ]], an open-source variable-capacity link emulator. Our goal is to be able to measure the sending rate of the video call app under investigation at any arbitrary time interval and compare it with the available bandwidth emulated by Cellsim during the same interval. Since their clocks are not synchronized, we need to start the sender and the emulators at the same time.
July 20, 2016, at 12:45 AM EST by 184.152.50.28 -
Changed line 29 from:
In our emulations, we used [[ https://github.com/keithw/multisend/archive/evaluated.zip | Cellsim ]], an open-source variable-capacity link emulator. You '''really''' should check out their main project [[http://alfalfa.mit.edu | Sprout ]] as well. We wanted to be able to compare Rebera
to:
In our emulations, we used [[ https://github.com/keithw/multisend/archive/evaluated.zip | Cellsim ]], an open-source variable-capacity link emulator. You '''really''' should check out their main project [[http://alfalfa.mit.edu | Sprout]] as well. We wanted to be able to compare Rebera
July 20, 2016, at 12:40 AM EST by 184.152.50.28 -
Changed line 29 from:
In our emulations, we used [[ https://github.com/keithw/multisend/archive/evaluated.zip | Cellsim ]], an open-source variable-capacity link emulator. You '''really''' should check out their main project [[http://bit.ly/2a9GMqA | Sprout ]] as well. We wanted to be able to compare Rebera
to:
In our emulations, we used [[ https://github.com/keithw/multisend/archive/evaluated.zip | Cellsim ]], an open-source variable-capacity link emulator. You '''really''' should check out their main project [[http://alfalfa.mit.edu | Sprout ]] as well. We wanted to be able to compare Rebera
July 20, 2016, at 12:33 AM EST by 184.152.50.28 -
Changed lines 19-20 from:
!!Reproducing our results
to:
!! Running Rebera
Changed lines 27-29 from:
!!Methodology for emulations

In our emulations, we used [[ https://github.com/keithw/multisend/archive/evaluated.zip | Cellsim ]], an open-source variable-capacity link emulator (you might want to check out their [[ http://alfalfa.mit.edu/ | project page ]] as well). We wanted to (to be continued)
to:
!! Methodology for emulations

In our emulations, we used [[ https://github.com/keithw/multisend/archive/evaluated.zip | Cellsim ]], an open-source variable-capacity link emulator. You '''really''' should check out their main project [[http://bit.ly/2a9GMqA | Sprout ]] as well. We wanted to be able to compare Rebera
July 19, 2016, at 09:13 PM EST by 172.16.9.58 -
Changed lines 24-28 from:
We specifically left the x264 library out, because the source code needs to be modified so as to generate an AVC stream with hierarchical-P coding structure with 3 temporal layers. You can download the source code [[ http://bit.ly/299X9wU | here]]. To find out how we modify the source code, you can follow [[ https://eymenkurdoglu.github.io/2016/07/01/hierp-one.html | this blog post ]]. Then you can configure, make and install as usual.
to:
We specifically left the x264 library out, because the source code needs to be modified so as to generate an AVC stream with hierarchical-P coding structure with 3 temporal layers. You can download the source code [[ http://bit.ly/299X9wU | here]]. To find out how we modify the source code, you can follow [[ https://eymenkurdoglu.github.io/2016/07/01/hierp-one.html | this blog post ]]. Then you can configure, make and install as usual.

!!Methodology for emulations

In our emulations, we used [[ https://github.com/keithw/multisend/archive/evaluated.zip | Cellsim ]], an open-source variable-capacity link emulator (you might want to check out their [[ http://alfalfa.mit.edu/ | project page ]] as well). We wanted to (to be continued)
July 08, 2016, at 01:38 AM EST by 184.152.50.28 -
Changed lines 14-18 from:
'''See the presentation slides'''
[[ http://vision.poly.edu/papers/2016/rebera_slides.pdf | [slides] ]]

'''Download the source code'''
Get the code [[ https://github.com/eymenkurdoglu/rebera | here ]].
to:
'''See the presentation [[ http://vision.poly.edu/papers/2016/rebera_slides.pdf | slides ]]'''

'''Download [[ https://github.com/eymenkurdoglu/rebera | Rebera's source code ]]'''
July 08, 2016, at 01:36 AM EST by 184.152.50.28 -
Changed lines 11-13 from:
'''Related Publication'''
* E. Kurdoglu, Y. Liu, Y. Wang, Y. Shi, C. Gu, J. Lyu, [[http://vision.poly.edu/papers/2016/a12-kurdoglu.pdf | "Real-time Bandwidth Prediction and Rate Estimation for Video Calls over Cellular Networks," ]] in Proc. of ACM MMSys, 2016 / Klagenfurt, Austria [[ http://vision.poly.edu/papers/2016/rebera_slides.pdf | [slides] ]]
to:
'''Read our paper'''
* E. Kurdoglu, Y. Liu, Y. Wang, Y. Shi, C. Gu, J. Lyu, [[http://vision.poly.edu/papers/2016/a12-kurdoglu.pdf | "Real-time Bandwidth Prediction and Rate Estimation for Video Calls over Cellular Networks," ]] in Proc. of ACM MMSys, 2016 / Klagenfurt, Austria

'''See the presentation slides'''
[[ http://vision.poly.edu/papers/2016/rebera_slides.pdf | [slides] ]]

'''Download the source code'''
Get the code [[ https://github.com/eymenkurdoglu/rebera | here ]].
Deleted lines 20-21:

Get Rebera's source code [[ https://github.com/eymenkurdoglu/rebera | here ]].
July 08, 2016, at 01:27 AM EST by 184.152.50.28 -
Added lines 15-16:

Get Rebera's source code [[ https://github.com/eymenkurdoglu/rebera | here ]].
July 07, 2016, at 11:37 AM EST by 172.16.10.20 -
Changed lines 1-2 from:
Please follow this link: https://github.com/eymenkurdoglu/rebera
to:
(:title Rebera: Real-time Bandwidth Estimation and Rate Adaptation for Video Calls over Cellular Networks:)

In this project we study interactive video calls between two users, where at least one of the users is connected over a cellular network. It is known that cellular links present highly-varying network bandwidth and packet delays. If the sending rate of the video call exceeds the available bandwidth, the video frames may be excessively delayed, destroying the interactivity of the video call.

In this study, we present Rebera, a cross-layer design of proactive congestion control, video encoding and rate adaptation, to maximize the video transmission rate while keeping the one-way frame delays sufficiently low. Rebera actively measures the available bandwidth in real-time by employing the video frames as packet trains. Using an online linear adaptive filter, Rebera makes a history-based prediction of the future capacity, and determines a bit budget for the video rate adaptation. Rebera uses the hierarchical-P video encoding structure to provide error resilience and to ease rate adaptation, while maintaining low encoding complexity and delay. Furthermore, Rebera decides in real time whether to send or discard an encoded frame, according to the budget, thereby preventing self-congestion and minimizing the packet delays. Our experiments with real cellular link traces demonstrate Rebera can, on average, deliver higher bandwidth utilization and shorter packet delays than Apple's FaceTime.

This research was done in collaboration with Tencent Inc. ([[http://www.wechat.com/en/ | WeChat International]]).
Changed lines 10-18 from:
* E. Kurdoglu, Y. Liu, Y. Wang, Y. Shi, C. Gu, J. Lyu, [[http://vision.poly.edu/papers/2016/a12-kurdoglu.pdf | "Real-time Bandwidth Prediction and Rate Estimation for Video Calls over Cellular Networks," ]] in Proc. of ACM MMSys, 2016 / Klagenfurt, Austria
to:
* E. Kurdoglu, Y. Liu, Y. Wang, Y. Shi, C. Gu, J. Lyu, [[http://vision.poly.edu/papers/2016/a12-kurdoglu.pdf | "Real-time Bandwidth Prediction and Rate Estimation for Video Calls over Cellular Networks," ]] in Proc. of ACM MMSys, 2016 / Klagenfurt, Austria

!!Reproducing our results

Rebera runs on Linux and uses [[http://www.videolan.org/developers/x264.html | x264]] library for H.264/AVC video encoding, [[http://ffmpeg.org/ | FFmpeg]] library for decoding, and [[https://www.libsdl.org/download-2.0.php | SDL2.0]] library for video display purposes, so you need to download these first.

@@sudo apt-get install libavformat-dev libavcodec-dev libavutil-dev libsdl2-dev@@

We specifically left the x264 library out, because the source code needs to be modified so as to generate an AVC stream with hierarchical-P coding structure with 3 temporal layers. You can download the source code [[ http://bit.ly/299X9wU | here]]. To find out how we modify the source code, you can follow [[ https://eymenkurdoglu.github.io/2016/07/01/hierp-one.html | this blog post ]]. Then you can configure, make and install as usual.
July 02, 2016, at 11:34 PM EST by 184.152.50.28 -
Changed lines 1-8 from:
(:title Rebera: Real-time Bandwidth Estimation and Rate Adaptation for Video Calls over Cellular Networks:)

In this project we study interactive video calls between two users, where at least one of the users is connected over a cellular network. It is known that cellular links present highly-varying network bandwidth and packet delays. If the sending rate of the video call exceeds the available bandwidth, the video frames may be excessively delayed, destroying the interactivity of the video call.

In this study, we present Rebera, a cross-layer design of proactive congestion control, video encoding and rate adaptation, to maximize the video transmission rate while keeping the one-way frame delays sufficiently low. Rebera actively measures the available bandwidth in real-time by employing the video frames as packet trains. Using an online linear adaptive filter, Rebera makes a history-based prediction of the future capacity, and determines a bit budget for the video rate adaptation. Rebera uses the hierarchical-P video encoding structure to provide error resilience and to ease rate adaptation, while maintaining low encoding complexity and delay. Furthermore, Rebera decides in real time whether to send or discard an encoded frame, according to the budget, thereby preventing self-congestion and minimizing the packet delays. Our experiments with real cellular link traces demonstrate Rebera can, on average, deliver higher bandwidth utilization and shorter packet delays than Apple's FaceTime.

This research was done in collaboration with Tencent Inc. ([[http://www.wechat.com/en/ | WeChat International]]).
to:
Please follow this link: https://github.com/eymenkurdoglu/rebera
Changed lines 4-10 from:
* E. Kurdoglu, Y. Liu, Y. Wang, Y. Shi, C. Gu, J. Lyu, [[http://vision.poly.edu/papers/2016/a12-kurdoglu.pdf | "Real-time Bandwidth Prediction and Rate Estimation for Video Calls over Cellular Networks," ]] in Proc. of ACM MMSys, 2016 / Klagenfurt, Austria

!!Reproducing our results

Rebera runs only on Linux and uses [[http://www.videolan.org/developers/x264.html | x264]] library for H.264/AVC video encoding, [[http://ffmpeg.org/ | ffmpeg]] library for decoding, and [[https://www.libsdl.org/download-2.0.php | SDL2.0]] library for video display purposes, so you need to download these first. The x264 library is modified to generate an AVC stream with hierarchical-P coding structure with 3 temporal layers, click [[https://wp.nyu.edu/eymen/2015/08/31/ | here]] to find out how.

Download [[http://www.example.com | Rebera]] and follow the configure-make routine.
to:
* E. Kurdoglu, Y. Liu, Y. Wang, Y. Shi, C. Gu, J. Lyu, [[http://vision.poly.edu/papers/2016/a12-kurdoglu.pdf | "Real-time Bandwidth Prediction and Rate Estimation for Video Calls over Cellular Networks," ]] in Proc. of ACM MMSys, 2016 / Klagenfurt, Austria
June 24, 2016, at 03:46 PM EST by 172.16.13.113 -
Changed line 10 from:
* E. Kurdoglu, Y. Liu, Y. Wang, Y. Shi, C. Gu, J. Lyu, "Real-time Bandwidth Prediction and Rate Estimation for Video Calls over Cellular Networks", in Proc. of ACM MMSys, 2016 / Klagenfurt, Austria
to:
* E. Kurdoglu, Y. Liu, Y. Wang, Y. Shi, C. Gu, J. Lyu, [[http://vision.poly.edu/papers/2016/a12-kurdoglu.pdf | Real-time Bandwidth Prediction and Rate Estimation for Video Calls over Cellular Networks," ]] in Proc. of ACM MMSys, 2016 / Klagenfurt, Austria
April 13, 2016, at 05:30 PM EST by Eymen Kurdoglu -
Changed lines 14-16 from:
Download [[http://www.example.com | Rebera]] and follow the configure-make routine. Rebera runs only on Linux and uses [[http://www.videolan.org/developers/x264.html | x264]] library for H.264/AVC video encoding, [[http://ffmpeg.org/ | ffmpeg]] library for decoding, and [[https://www.libsdl.org/download-2.0.php | SDL2.0]] library for video display purposes. The x264 library is modified to generate an AVC stream with hierarchical-P coding structure with 3 temporal layers, click [[https://wp.nyu.edu/eymen/2015/08/31/ | here]] to find out how.
to:
Rebera runs only on Linux and uses [[http://www.videolan.org/developers/x264.html | x264]] library for H.264/AVC video encoding, [[http://ffmpeg.org/ | ffmpeg]] library for decoding, and [[https://www.libsdl.org/download-2.0.php | SDL2.0]] library for video display purposes, so you need to download these first. The x264 library is modified to generate an AVC stream with hierarchical-P coding structure with 3 temporal layers, click [[https://wp.nyu.edu/eymen/2015/08/31/ | here]] to find out how.

Download [[http://www.example.com | Rebera]] and follow the configure-make routine.
April 13, 2016, at 05:15 PM EST by Eymen Kurdoglu -
Added lines 6-7:

This research was done in collaboration with Tencent Inc. ([[http://www.wechat.com/en/ | WeChat International]]).
April 13, 2016, at 04:54 PM EST by Eymen Kurdoglu -
Changed lines 10-12 from:
!!Reproducing our results!!

Download Rebera [[http://www.example.com | here]]. Rebera uses [[http://www.videolan.org/developers/x264.html | x264]] library for H.264/AVC video encoding, [[http://ffmpeg.org/ | ffmpeg]] library for decoding, and [[https://www.libsdl.org/download-2.0.php | SDL2.0]] library for display purposes. The x264 library is modified to generate an AVC stream with hierarchical-P coding structure with 3 temporal layers, click [[https://wp.nyu.edu/eymen/2015/08/31/ | here]] to find out how.
to:
!!Reproducing our results

Download Rebera [[http://www.example.com | here]] and follow the configure-make routine. Rebera runs only on Linux and uses [[http://www.videolan.org/developers/x264.html | x264]] library for H.264/AVC video encoding, [[http://ffmpeg.org/ | ffmpeg]] library for decoding, and [[https://www.libsdl.org/download-2.0.php | SDL2.0]] library for video display purposes. The x264 library is modified to generate an AVC stream with hierarchical-P coding structure with 3 temporal layers, click [[https://wp.nyu.edu/eymen/2015/08/31/ | here]] to find out how.
April 13, 2016, at 04:49 PM EST by Eymen Kurdoglu - Description of the Rebera project
Changed line 12 from:
Download Rebera [[http://www.example.com | here]]. Rebera uses [[http://www.videolan.org/developers/x264.html | x264]] library for H.264/AVC video encoding, [[http://ffmpeg.org/ | ffmpeg]] library for decoding, and [[https://www.libsdl.org/download-2.0.php | SDL2.0]] library for display purposes. The x264 library is modified to generate an AVC stream with hierarchical-P coding structure, click [[https://wp.nyu.edu/eymen/2015/08/31/ | here]] to find out how.
to:
Download Rebera [[http://www.example.com | here]]. Rebera uses [[http://www.videolan.org/developers/x264.html | x264]] library for H.264/AVC video encoding, [[http://ffmpeg.org/ | ffmpeg]] library for decoding, and [[https://www.libsdl.org/download-2.0.php | SDL2.0]] library for display purposes. The x264 library is modified to generate an AVC stream with hierarchical-P coding structure with 3 temporal layers, click [[https://wp.nyu.edu/eymen/2015/08/31/ | here]] to find out how.
April 13, 2016, at 04:47 PM EST by 172.16.15.103 -
Changed line 12 from:
Download Rebera [[http://www.example.com | here]]. We have used [[http://www.videolan.org/developers/x264.html | x264]] library for H.264/AVC video encoding, [[http://ffmpeg.org/ | ffmpeg]] library for decoding, and [[https://www.libsdl.org/download-2.0.php | SDL2.0]] library for display purposes.
to:
Download Rebera [[http://www.example.com | here]]. Rebera uses [[http://www.videolan.org/developers/x264.html | x264]] library for H.264/AVC video encoding, [[http://ffmpeg.org/ | ffmpeg]] library for decoding, and [[https://www.libsdl.org/download-2.0.php | SDL2.0]] library for display purposes. The x264 library is modified to generate an AVC stream with hierarchical-P coding structure, click [[https://wp.nyu.edu/eymen/2015/08/31/ | here]] to find out how.
April 13, 2016, at 04:43 PM EST by 172.16.15.103 -
Changed line 12 from:
Download Rebera [[here]].
to:
Download Rebera [[http://www.example.com | here]]. We have used [[http://www.videolan.org/developers/x264.html | x264]] library for H.264/AVC video encoding, [[http://ffmpeg.org/ | ffmpeg]] library for decoding, and [[https://www.libsdl.org/download-2.0.php | SDL2.0]] library for display purposes.
April 13, 2016, at 04:39 PM EST by 172.16.15.103 -
Changed lines 8-12 from:
* E. Kurdoglu, Y. Liu, Y. Wang, Y. Shi, C. Gu, J. Lyu, "Real-time Bandwidth Prediction and Rate Estimation for Video Calls over Cellular Networks", in Proc. of ACM MMSys, 2016 / Klagenfurt, Austria
to:
* E. Kurdoglu, Y. Liu, Y. Wang, Y. Shi, C. Gu, J. Lyu, "Real-time Bandwidth Prediction and Rate Estimation for Video Calls over Cellular Networks", in Proc. of ACM MMSys, 2016 / Klagenfurt, Austria

!!Reproducing our results!!

Download Rebera [[here]].
April 13, 2016, at 04:34 PM EST by 172.16.15.103 -
Added lines 1-2:
(:title Rebera: Real-time Bandwidth Estimation and Rate Adaptation for Video Calls over Cellular Networks:)
April 13, 2016, at 04:32 PM EST by 172.16.15.103 -
Changed lines 1-6 from:
Lorem ipsum
to:
In this project we study interactive video calls between two users, where at least one of the users is connected over a cellular network. It is known that cellular links present highly-varying network bandwidth and packet delays. If the sending rate of the video call exceeds the available bandwidth, the video frames may be excessively delayed, destroying the interactivity of the video call.

In this study, we present Rebera, a cross-layer design of proactive congestion control, video encoding and rate adaptation, to maximize the video transmission rate while keeping the one-way frame delays sufficiently low. Rebera actively measures the available bandwidth in real-time by employing the video frames as packet trains. Using an online linear adaptive filter, Rebera makes a history-based prediction of the future capacity, and determines a bit budget for the video rate adaptation. Rebera uses the hierarchical-P video encoding structure to provide error resilience and to ease rate adaptation, while maintaining low encoding complexity and delay. Furthermore, Rebera decides in real time whether to send or discard an encoded frame, according to the budget, thereby preventing self-congestion and minimizing the packet delays. Our experiments with real cellular link traces demonstrate Rebera can, on average, deliver higher bandwidth utilization and shorter packet delays than Apple's FaceTime.

'''Related Publication'''
* E. Kurdoglu, Y. Liu, Y. Wang, Y. Shi, C. Gu, J. Lyu, "Real-time Bandwidth Prediction and Rate Estimation for Video Calls over Cellular Networks", in Proc. of ACM MMSys, 2016 / Klagenfurt, Austria
February 12, 2016, at 04:58 PM EST by 172.16.11.216 -
Added line 1:
Lorem ipsum
Page last modified on August 08, 2016, at 02:38 PM EST