Media‐Aware
Network
Elements

                     on
Legacy
Devices


                                           m16695


                Ingo
Kofler,
Robert
Kuschnig,
and
Hermann
Hellwagner

                                 Chris:an
Timmerer



                 Klagenfurt
University
(UNIKLU)

Faculty
of
Technical
Sciences
(TEWI)

          Department
of
Informa:on
Technology
(ITEC)

Mul:media
Communica:on
(MMC)

 h;p://research.@mmerer.com

h;p://blog.@mmerer.com

mailto:chris@an.@mmerer@itec.uni‐klu.ac.at



Acknowledgement:
Part
of
this
work
is
supported
by
the
European
Commission
in
the

 context
of
the
and
ENTHRONE
(contract
no.
038463)
project.
Further
informa@on
is

                      available
at
h;p://www.ist‐enthrone.org.


Outline

•  Mo@va@on
and
Introduc@on

•  List
of
Technologies


•  Architecture
and
Performance
Evalua@ons


•  Demo
Video


•  Conclusions
/
References



2009/07/01
      Chris@an
Timmerer,
Klagenfurt
University,
Austria
   2

Mo@va@on
and
Introduc@on

•  Adapta@on
of
an
SVC
bitstream

      –  Achieved
by
removing
certain
NALUs
➙
filtering
of
NALUs

      –  Steered
by
a
(TID,
DID,
QID)
tuple
➙
filter
criteria

      –  Computa@onally
cheap
(compared
to
transcoding
etc.)


•  Idea

      –  Perform
real‐@me
in‐network
adapta@on
of
the
SVC
bitstream

         on
an
ordinary,
low‐cost
WiFi
router

      –  Media‐aware
Network
Element
(MANE)


•  Applica@ons
of
in‐network
adapta@on

      –  Cross‐layer
adapta@on
on
the
access
point

      –  Adapta@on
for
different
end‐devices

2009/07/01
            Chris@an
Timmerer,
Klagenfurt
University,
Austria
   3

Media‐aware
Network
Element





              On‐the‐fly
adapta@on
of
scalable
coded
video
content
in
a
media‐
                             aware
network
element
(MANE).




                                                 Source:
h;p://ip.hhi.de/imagecom_G1/savce/

2009/07/01
                 Chris@an
Timmerer,
Klagenfurt
University,
Austria
            4

List
of
Technologies

•  Scalable
Video
Coding
(SVC)

•  Real‐@me
Streaming
Protocol
(RTSP)

      –  Establishing
and
controlling
the
streaming
session

      –  VCR‐like
control
of
the
streaming
(Start,
Stop,
Pause)

•  Real‐@me
Transport
Protocol
(RTP)

      –  Encapsulates
the
video
and/or
audio
content

      –  Mostly
used
on
top
of
the
unreliable
UDP
protocol

      –  Offers
sequence
number,
@mestamps
for
syncing
the

         playback

      –  Generic
header
with
content‐specific
payload
format

         (AVC,
SVC,
…)


2009/07/01
           Chris@an
Timmerer,
Klagenfurt
University,
Austria
   5

Proxy Approach
              server
                router
with
proxy

                                          router
                                           client


      RTSP
      RTP/RTCP
             RTSP
      RTP/RTCP
                         RTSP
     RTP/RTCP


      TCP
            UDP
              TCP
           UDP
                         TCP
           UDP


                IP
                              IP
                                         IP


          MAC/PHY
                          MAC/PHY
                                    MAC/PHY



      •  Proxy on network device between client & server
              –  Intercepts the RTSP / RTP communication
              –  Proxy is transparent for the client
              –  Acts as client for initial server and as server for the client
      •  Implications
              –  Proxy has to modify parts of the request (e.g. port numbers)
              –  Proxy can then adapt the SVC video stream carried over RTP
2009/07/01
                    Chris@an
Timmerer,
Klagenfurt
University,
Austria
                         6

Proxy Approach (cont’d)

Benefits
                                       Drawbacks

•  Session‐
and
stream‐/                       •  Proxy
is
running
as
user‐
   media‐awareness

                              space
process


•  Enables
stateful
inspec@on
                 •  RTP
packets
have
to
be

   and
processing
of
packets
                     passed
(copied)
from
the

•  Allows
adapta@on
on
a
per‐                     kernel‐space
to
the
user‐
   session
basis
                                 space
and
vice
versa

•  Consistent
RTCP
receiver
                   •  Decreases
theore@cal

   and
sender
reports
                            throughput
compared
to

                                                  kernel‐internal
solu@on
(e.g.

                                                  filtering
on
IP
level)


2009/07/01
         Chris@an
Timmerer,
Klagenfurt
University,
Austria
        7

Architecture

                                                    Handles
incoming
RTSP
requests
(554

                                                    redirect
to
proxy
using
iptables)


                                                    SDP
stored
in
LRU‐based
cache
un@l

                                                    session
established


                                                    Maintains
state:
seq#,
@mestamps,

                                                    SVC
params,
monitoring,
etc.


                                                    Forwards
non‐SVC
packets

                                                    Adapts
SVC
packets


                                                    Acts
as
server
for
client

                                                    Acts
as
client
for
server


                                                    Retrieve
monitoring
informa@on

                                                    Modify
adapta@on/SVC
parameters


2009/07/01
   Chris@an
Timmerer,
Klagenfurt
University,
Austria
                    8

Performance
Evalua@ons

•  Hardware:
Linksys
WRT
54
GL

      –  Broadcom
System‐on‐Chip
BCM5352EL

      –  MIPS32
200
MHz
CPU

      –  16
MB
RAM

      –  4
MB
Flash
Memory

      –  IEEE
802.11b/g
WLAN

      –  Fast
Ethernet
switch
with
5
ports

      –  price
~
45
Euros

(May
2008)

•  Sonware:
OpenWrt

      –  Linux‐based
firmware
for
Broadcom‐based
WiFi
routers

      –  gcc-based SDK for OpenWrt available
      –  Proxy
implementa@on
in
ANSI
C

2009/07/01
          Chris@an
Timmerer,
Klagenfurt
University,
Austria
   9

Performance
Evalua@ons
(cont’d)

•  Evalua@on
of
the
implementa@on
on
the
target
plaporm

      –  Server
–
proxy
–
client
deployment

      –  Inves@ga@on
of
worst‐case
scenario
(no
adapta@on)


•  Performance
metrics

      –  CPU
usage
for
different
number
of
streams

      –  Delay
introduced
by
the
proxy
on
a
complete
access
unit
(frame)


•  Three
different
SVC
streams
for
evalua@on

      –  Foreman,
CIF,
30
Hz,
715
kbps

      –  City,
4CIF,
30
Hz,
1247
kbps

      –  Harbour,
4CIF,
30
Hz,
2056
kbps


2009/07/01
             Chris@an
Timmerer,
Klagenfurt
University,
Austria
   10

Performance
Evalua@ons
(cont’d)

CPU
Usage





2009/07/01
    Chris@an
Timmerer,
Klagenfurt
University,
Austria
   11

Performance
Evalua@ons
(cont’d)

CDF of delay for sequence city (1245 kbps)





2009/07/01
      Chris@an
Timmerer,
Klagenfurt
University,
Austria
   12

Demo
Video





2009/07/01
   Chris@an
Timmerer,
Klagenfurt
University,
Austria
   13

Conclusions

•  Proxy
approach
for
in‐network
adapta@on
on
a
per‐packet

   basis

      –  Aware
of
sessions
and
individual
streams
(media‐aware)

      –  Stateful,
not
a
simple
packet
dropper

•  Applica@ons

      –  Adapta@on
according
to
device
capabili@es

      –  Cross‐layer
adapta@on

      –  NAT
traversal

•  Performance
sufficient
for
typical
home
deployments

      –  4
parallel
streams
with
30
percent
CPU
load
and
<
100
ms
delay



➙SVC
adapta@on
can
be
done
on
exis@ng
off‐the‐shelf

  network
devices

2009/07/01
            Chris@an
Timmerer,
Klagenfurt
University,
Austria
   14

References

•  I.
Kofler,
J.
Seidl,
C.
Timmerer,
H.
Hellwagner,
I.
Djama
and
T.
Ahmed,

   “Using
MPEG‐21
for
cross‐layer
mul@media
content
adapta@on”,
Journal

   on
Signal,
Image
and
Video
Processing,
Springer,
vol.
2,
no.
4,
Dec.
2008.

•  R.
Kuschnig,
I.
Kofler,
M.
Ransburg,
H.
Hellwagner,
“Design
op@ons
and

   comparison
of
in‐network
H.264/SVC
adapta@on”,
Journal
of
Visual

   Communica@on
and
Image
Representa@on,
Sept.
2008.

•  I.
Kofler,
M.
Prangl,
R.
Kuschnig,
and
H.
Hellwagner,
“An
H.264/SVC‐based

   adapta@on
proxy
on
a
WiFi
router”,
Proceedings
of
the
18th
Interna@onal

   Workshop
on
Network
and
Opera@ng
Systems
Support
for
Digital
Audio

   and
Video
(NOSSDAV
2008),
Braunschweig,
Germany,
May
2008.

•  I.
Kofler,
C.
Timmerer,
H.
Hellwagner,
and
T.
Ahmed:
“Towards
MPEG‐21‐
   based
Cross‐layer
Mul@media
Content
Adapta@on”,
Proc.
2nd

   Interna@onal
Workshop
on
Seman@c
Media
Adapta@on
and

   Personaliza@on
(SMAP
2007),
London,
UK,
Dec.
2007.





2009/07/01
             Chris@an
Timmerer,
Klagenfurt
University,
Austria
   15

Thank
you
for
your
a;en@on



              ...
ques@ons,
comments,
etc.
are
welcome
…





                                                            
Ass.‐Prof.
Dipl.‐Ing.
Dr.
Chris@an
Timmerer

                                   Klagenfurt
University,
Department
of
Informa@on
Technology
(ITEC)

                                                Universitätsstrasse
65‐67,
A‐9020
Klagenfurt,
AUSTRIA

                                                                  chris@an.@mmerer@itec.uni‐klu.ac.at

                                                                         h;p://research.@mmerer.com/

                                                     Tel:
+43/463/2700
3621
Fax:
+43/463/2700
3699

                                                                                  ©
Copyright:
Chris.an
Timmerer





2009/07/01
              Chris@an
Timmerer,
Klagenfurt
University,
Austria
                                         16


More Related Content

PPTX
PDF
Technology Update: MPEG-Dash
PPTX
HTTP Streaming of MPEG Media
PDF
LwTE: Light-weight Transcoding at the Edge
PDF
Motion Vector Recovery for Real-time H.264 Video Streams
PPTX
DASH at the ACM Multimedia 2011
PPTX
Labmeeting - 20150831 - Overhead and Performance of Low Latency Live Streamin...
PPTX
Using SVC for DASH in Mobile Environments
Technology Update: MPEG-Dash
HTTP Streaming of MPEG Media
LwTE: Light-weight Transcoding at the Edge
Motion Vector Recovery for Real-time H.264 Video Streams
DASH at the ACM Multimedia 2011
Labmeeting - 20150831 - Overhead and Performance of Low Latency Live Streamin...
Using SVC for DASH in Mobile Environments

What's hot (20)

PPTX
CAdViSE or how to find the Sweet Spots of ABR Systems
PDF
MiPSO: Multi-Period Per-Scene Optimization For HTTP Adaptive Streaming
PDF
An Introduction to OMNeT++ 5.4
PDF
20 Years of Streaming in 20 Minutes
PPTX
Scalable Video Coding Guidelines and Performance Evaluations for Adaptive Me...
PPTX
Bandwidth Prediction in Low-Latency Chunked Streaming
PDF
CAdViSE: Cloud based Adaptive Video Streaming Evaluation Framework for the Au...
PPTX
HTTP Adaptive Streaming State of the Art and Challenges Ahead
PDF
FAUST: Fast Per-Scene Encoding Using Entropy-Based Scene Detection and Machin...
PPTX
Dynamic Adaptive Point Cloud Streaming
PDF
ES-HAS: An Edge- and SDN-Assisted Framework for HTTP Adaptive Video Streaming
PDF
HTTP Adaptive Streaming – Quo Vadis?
PPT
Video Quality Measurements
PPTX
Adaptive Video over ICN @ IETF'87
PPTX
On Optimizing Resource Utilization in AVC-based Real-time Video Streaming
PDF
A Distributed Approach for Bitrate Selection in HTTP Adaptive Streaming
PPTX
Dynamic Adaptive Streaming over HTTP (DASH)
PDF
HTTP Adaptive Streaming – Where Is It Heading?
CAdViSE or how to find the Sweet Spots of ABR Systems
MiPSO: Multi-Period Per-Scene Optimization For HTTP Adaptive Streaming
An Introduction to OMNeT++ 5.4
20 Years of Streaming in 20 Minutes
Scalable Video Coding Guidelines and Performance Evaluations for Adaptive Me...
Bandwidth Prediction in Low-Latency Chunked Streaming
CAdViSE: Cloud based Adaptive Video Streaming Evaluation Framework for the Au...
HTTP Adaptive Streaming State of the Art and Challenges Ahead
FAUST: Fast Per-Scene Encoding Using Entropy-Based Scene Detection and Machin...
Dynamic Adaptive Point Cloud Streaming
ES-HAS: An Edge- and SDN-Assisted Framework for HTTP Adaptive Video Streaming
HTTP Adaptive Streaming – Quo Vadis?
Video Quality Measurements
Adaptive Video over ICN @ IETF'87
On Optimizing Resource Utilization in AVC-based Real-time Video Streaming
A Distributed Approach for Bitrate Selection in HTTP Adaptive Streaming
Dynamic Adaptive Streaming over HTTP (DASH)
HTTP Adaptive Streaming – Where Is It Heading?
Ad

Viewers also liked (7)

PDF
mHealth Summit EU 2015
PPTX
Diapositivas capitulo 4
PDF
Center for connected medicine the surgeon can see you now
PPS
Votado Como El Mejor Mail Del AñO (2)
PDF
David Doherty presentation on mHealth Best Practice at Doctors 2.0 and You 2014
PDF
Pericia química legal
mHealth Summit EU 2015
Diapositivas capitulo 4
Center for connected medicine the surgeon can see you now
Votado Como El Mejor Mail Del AñO (2)
David Doherty presentation on mHealth Best Practice at Doctors 2.0 and You 2014
Pericia química legal
Ad

Similar to Media-Aware Network Elements on Legacy Devices (20)

PPT
Optimizing User QoE through Overlay Routing, Bandwidth ...
PDF
Pre-Con Education: Recognizing Your Network's Key Performance Indicators Th...
PPTX
06-dash.pptx
PDF
network basics
PDF
2002023
PDF
Delivering Traditional and Omnidirectional Media
PDF
Uit Presentation of IN/NGIN for Cosmote 2010
PDF
ROLE OF DIGITAL SIMULATION IN CONFIGURING NETWORK PARAMETERS
DOC
Prashant Resume
PPTX
Globecom 2015: Adaptive Raptor Carousel for 802.11
PDF
2018 FRSecure CISSP Mentor Program- Session 7
PPT
Providing Controlled Quality Assurance in Video Streaming ...
PPTX
MPEG-DASH open source tools and cloud services
PDF
CloudLightning and the OPM-based Use Case
DOC
[32]
PDF
ADVANCES IN CHANNEL-ADAPTIVE VIDEO STREAMING
PDF
WebRTC Real time media P2P, Server, Infrastructure, and Platform
PDF
Multicasting Of Adaptively-Encoded MPEG4 Over Qos-Cognizant IP Networks
PPTX
A QoS-Adaptive Framework for Screen Sharing Over Internet
PPTX
2013 09-02 senzations-bimschas-part1-smart-santander-facility-luebeck
Optimizing User QoE through Overlay Routing, Bandwidth ...
Pre-Con Education: Recognizing Your Network's Key Performance Indicators Th...
06-dash.pptx
network basics
2002023
Delivering Traditional and Omnidirectional Media
Uit Presentation of IN/NGIN for Cosmote 2010
ROLE OF DIGITAL SIMULATION IN CONFIGURING NETWORK PARAMETERS
Prashant Resume
Globecom 2015: Adaptive Raptor Carousel for 802.11
2018 FRSecure CISSP Mentor Program- Session 7
Providing Controlled Quality Assurance in Video Streaming ...
MPEG-DASH open source tools and cloud services
CloudLightning and the OPM-based Use Case
[32]
ADVANCES IN CHANNEL-ADAPTIVE VIDEO STREAMING
WebRTC Real time media P2P, Server, Infrastructure, and Platform
Multicasting Of Adaptively-Encoded MPEG4 Over Qos-Cognizant IP Networks
A QoS-Adaptive Framework for Screen Sharing Over Internet
2013 09-02 senzations-bimschas-part1-smart-santander-facility-luebeck

More from Alpen-Adria-Universität (20)

PDF
Energy-Quality-aware Variable Framerate Pareto-Front for Adaptive Video Strea...
PPTX
End-to-end Quality of Experience Evaluation for HTTP Adaptive Streaming
PDF
HTTP Adaptive Streaming – Quo Vadis (2024)
PDF
Energy Efficient Video Encoding for Cloud and Edge Computing Instances
PDF
Video Streaming: Then, Now, and in the Future
PDF
VEED: Video Encoding Energy and CO2 Emissions Dataset for AWS EC2 instances
PDF
GREEM: An Open-Source Energy Measurement Tool for Video Processing
PDF
Optimal Quality and Efficiency in Adaptive Live Streaming with JND-Aware Low ...
PDF
VEEP: Video Encoding Energy and CO₂ Emission Prediction
PDF
Content-adaptive Video Coding for HTTP Adaptive Streaming
PPTX
Empowerment of Atypical Viewers via Low-Effort Personalized Modeling of Video...
PPTX
Empowerment of Atypical Viewers via Low-Effort Personalized Modeling of Vid...
PPTX
Optimizing Video Streaming for Sustainability and Quality: The Role of Prese...
PDF
Energy-Efficient Multi-Codec Bitrate-Ladder Estimation for Adaptive Video Str...
PPTX
Machine Learning Based Resource Utilization Prediction in the Computing Conti...
PDF
Evaluation of Quality of Experience of ABR Schemes in Gaming Stream
PDF
Network-Assisted Delivery of Adaptive Video Streaming Services through CDN, S...
PDF
Multi-access Edge Computing for Adaptive Video Streaming
PPTX
Policy-Driven Dynamic HTTP Adaptive Streaming Player Environment
PDF
VE-Match: Video Encoding Matching-based Model for Cloud and Edge Computing In...
Energy-Quality-aware Variable Framerate Pareto-Front for Adaptive Video Strea...
End-to-end Quality of Experience Evaluation for HTTP Adaptive Streaming
HTTP Adaptive Streaming – Quo Vadis (2024)
Energy Efficient Video Encoding for Cloud and Edge Computing Instances
Video Streaming: Then, Now, and in the Future
VEED: Video Encoding Energy and CO2 Emissions Dataset for AWS EC2 instances
GREEM: An Open-Source Energy Measurement Tool for Video Processing
Optimal Quality and Efficiency in Adaptive Live Streaming with JND-Aware Low ...
VEEP: Video Encoding Energy and CO₂ Emission Prediction
Content-adaptive Video Coding for HTTP Adaptive Streaming
Empowerment of Atypical Viewers via Low-Effort Personalized Modeling of Video...
Empowerment of Atypical Viewers via Low-Effort Personalized Modeling of Vid...
Optimizing Video Streaming for Sustainability and Quality: The Role of Prese...
Energy-Efficient Multi-Codec Bitrate-Ladder Estimation for Adaptive Video Str...
Machine Learning Based Resource Utilization Prediction in the Computing Conti...
Evaluation of Quality of Experience of ABR Schemes in Gaming Stream
Network-Assisted Delivery of Adaptive Video Streaming Services through CDN, S...
Multi-access Edge Computing for Adaptive Video Streaming
Policy-Driven Dynamic HTTP Adaptive Streaming Player Environment
VE-Match: Video Encoding Matching-based Model for Cloud and Edge Computing In...

Recently uploaded (20)

PDF
sustainability-14-14877-v2.pddhzftheheeeee
PDF
A proposed approach for plagiarism detection in Myanmar Unicode text
PPTX
TEXTILE technology diploma scope and career opportunities
PDF
Five Habits of High-Impact Board Members
PDF
How IoT Sensor Integration in 2025 is Transforming Industries Worldwide
PDF
Produktkatalog für HOBO Datenlogger, Wetterstationen, Sensoren, Software und ...
PPT
What is a Computer? Input Devices /output devices
PDF
NewMind AI Weekly Chronicles – August ’25 Week III
PDF
How ambidextrous entrepreneurial leaders react to the artificial intelligence...
PPTX
The various Industrial Revolutions .pptx
PDF
OpenACC and Open Hackathons Monthly Highlights July 2025
DOCX
search engine optimization ppt fir known well about this
PPTX
Configure Apache Mutual Authentication
PDF
Architecture types and enterprise applications.pdf
PDF
Enhancing plagiarism detection using data pre-processing and machine learning...
PPTX
Module 1 Introduction to Web Programming .pptx
PDF
A contest of sentiment analysis: k-nearest neighbor versus neural network
PDF
The influence of sentiment analysis in enhancing early warning system model f...
PPTX
Microsoft Excel 365/2024 Beginner's training
DOCX
Basics of Cloud Computing - Cloud Ecosystem
sustainability-14-14877-v2.pddhzftheheeeee
A proposed approach for plagiarism detection in Myanmar Unicode text
TEXTILE technology diploma scope and career opportunities
Five Habits of High-Impact Board Members
How IoT Sensor Integration in 2025 is Transforming Industries Worldwide
Produktkatalog für HOBO Datenlogger, Wetterstationen, Sensoren, Software und ...
What is a Computer? Input Devices /output devices
NewMind AI Weekly Chronicles – August ’25 Week III
How ambidextrous entrepreneurial leaders react to the artificial intelligence...
The various Industrial Revolutions .pptx
OpenACC and Open Hackathons Monthly Highlights July 2025
search engine optimization ppt fir known well about this
Configure Apache Mutual Authentication
Architecture types and enterprise applications.pdf
Enhancing plagiarism detection using data pre-processing and machine learning...
Module 1 Introduction to Web Programming .pptx
A contest of sentiment analysis: k-nearest neighbor versus neural network
The influence of sentiment analysis in enhancing early warning system model f...
Microsoft Excel 365/2024 Beginner's training
Basics of Cloud Computing - Cloud Ecosystem

Media-Aware Network Elements on Legacy Devices

  • 1. Media‐Aware
Network
Elements
 on
Legacy
Devices

 m16695
 Ingo
Kofler,
Robert
Kuschnig,
and
Hermann
Hellwagner
 Chris:an
Timmerer
 Klagenfurt
University
(UNIKLU)

Faculty
of
Technical
Sciences
(TEWI)
 Department
of
Informa:on
Technology
(ITEC)

Mul:media
Communica:on
(MMC)
 h;p://[email protected]

h;p://[email protected]

mailto:chris@an.@[email protected]‐klu.ac.at
 Acknowledgement:
Part
of
this
work
is
supported
by
the
European
Commission
in
the
 context
of
the
and
ENTHRONE
(contract
no.
038463)
project.
Further
informa@on
is
 available
at
h;p://www.ist‐enthrone.org.


  • 2. Outline
 •  Mo@va@on
and
Introduc@on
 •  List
of
Technologies
 •  Architecture
and
Performance
Evalua@ons
 •  Demo
Video
 •  Conclusions
/
References
 2009/07/01
 Chris@an
Timmerer,
Klagenfurt
University,
Austria
 2

  • 3. Mo@va@on
and
Introduc@on
 •  Adapta@on
of
an
SVC
bitstream
 –  Achieved
by
removing
certain
NALUs
➙
filtering
of
NALUs
 –  Steered
by
a
(TID,
DID,
QID)
tuple
➙
filter
criteria
 –  Computa@onally
cheap
(compared
to
transcoding
etc.)
 •  Idea
 –  Perform
real‐@me
in‐network
adapta@on
of
the
SVC
bitstream
 on
an
ordinary,
low‐cost
WiFi
router
 –  Media‐aware
Network
Element
(MANE)
 •  Applica@ons
of
in‐network
adapta@on
 –  Cross‐layer
adapta@on
on
the
access
point
 –  Adapta@on
for
different
end‐devices
 2009/07/01
 Chris@an
Timmerer,
Klagenfurt
University,
Austria
 3

  • 4. Media‐aware
Network
Element
 On‐the‐fly
adapta@on
of
scalable
coded
video
content
in
a
media‐ aware
network
element
(MANE).
 Source:
h;p://ip.hhi.de/imagecom_G1/savce/
 2009/07/01
 Chris@an
Timmerer,
Klagenfurt
University,
Austria
 4

  • 5. List
of
Technologies
 •  Scalable
Video
Coding
(SVC)
 •  Real‐@me
Streaming
Protocol
(RTSP)
 –  Establishing
and
controlling
the
streaming
session
 –  VCR‐like
control
of
the
streaming
(Start,
Stop,
Pause)
 •  Real‐@me
Transport
Protocol
(RTP)
 –  Encapsulates
the
video
and/or
audio
content
 –  Mostly
used
on
top
of
the
unreliable
UDP
protocol
 –  Offers
sequence
number,
@mestamps
for
syncing
the
 playback
 –  Generic
header
with
content‐specific
payload
format
 (AVC,
SVC,
…)
 2009/07/01
 Chris@an
Timmerer,
Klagenfurt
University,
Austria
 5

  • 6. Proxy Approach server
 router
with
proxy
 router
 client
 RTSP
 RTP/RTCP
 RTSP
 RTP/RTCP
 RTSP
 RTP/RTCP
 TCP
 UDP
 TCP
 UDP
 TCP
 UDP
 IP
 IP
 IP
 MAC/PHY
 MAC/PHY
 MAC/PHY
 •  Proxy on network device between client & server –  Intercepts the RTSP / RTP communication –  Proxy is transparent for the client –  Acts as client for initial server and as server for the client •  Implications –  Proxy has to modify parts of the request (e.g. port numbers) –  Proxy can then adapt the SVC video stream carried over RTP 2009/07/01
 Chris@an
Timmerer,
Klagenfurt
University,
Austria
 6

  • 7. Proxy Approach (cont’d)
 Benefits
 Drawbacks
 •  Session‐
and
stream‐/ •  Proxy
is
running
as
user‐ media‐awareness

 space
process

 •  Enables
stateful
inspec@on
 •  RTP
packets
have
to
be
 and
processing
of
packets
 passed
(copied)
from
the
 •  Allows
adapta@on
on
a
per‐ kernel‐space
to
the
user‐ session
basis
 space
and
vice
versa
 •  Consistent
RTCP
receiver
 •  Decreases
theore@cal
 and
sender
reports
 throughput
compared
to
 kernel‐internal
solu@on
(e.g.
 filtering
on
IP
level)
 2009/07/01
 Chris@an
Timmerer,
Klagenfurt
University,
Austria
 7

  • 8. Architecture
 Handles
incoming
RTSP
requests
(554
 redirect
to
proxy
using
iptables)
 SDP
stored
in
LRU‐based
cache
un@l
 session
established
 Maintains
state:
seq#,
@mestamps,
 SVC
params,
monitoring,
etc.
 Forwards
non‐SVC
packets
 Adapts
SVC
packets
 Acts
as
server
for
client
 Acts
as
client
for
server
 Retrieve
monitoring
informa@on
 Modify
adapta@on/SVC
parameters
 2009/07/01
 Chris@an
Timmerer,
Klagenfurt
University,
Austria
 8

  • 9. Performance
Evalua@ons
 •  Hardware:
Linksys
WRT
54
GL
 –  Broadcom
System‐on‐Chip
BCM5352EL
 –  MIPS32
200
MHz
CPU
 –  16
MB
RAM
 –  4
MB
Flash
Memory
 –  IEEE
802.11b/g
WLAN
 –  Fast
Ethernet
switch
with
5
ports
 –  price
~
45
Euros

(May
2008)
 •  Sonware:
OpenWrt
 –  Linux‐based
firmware
for
Broadcom‐based
WiFi
routers
 –  gcc-based SDK for OpenWrt available –  Proxy
implementa@on
in
ANSI
C
 2009/07/01
 Chris@an
Timmerer,
Klagenfurt
University,
Austria
 9

  • 10. Performance
Evalua@ons
(cont’d)
 •  Evalua@on
of
the
implementa@on
on
the
target
plaporm
 –  Server
–
proxy
–
client
deployment
 –  Inves@ga@on
of
worst‐case
scenario
(no
adapta@on)
 •  Performance
metrics
 –  CPU
usage
for
different
number
of
streams
 –  Delay
introduced
by
the
proxy
on
a
complete
access
unit
(frame)
 •  Three
different
SVC
streams
for
evalua@on
 –  Foreman,
CIF,
30
Hz,
715
kbps
 –  City,
4CIF,
30
Hz,
1247
kbps
 –  Harbour,
4CIF,
30
Hz,
2056
kbps
 2009/07/01
 Chris@an
Timmerer,
Klagenfurt
University,
Austria
 10

  • 11. Performance
Evalua@ons
(cont’d)
 CPU
Usage
 2009/07/01
 Chris@an
Timmerer,
Klagenfurt
University,
Austria
 11

  • 12. Performance
Evalua@ons
(cont’d)
 CDF of delay for sequence city (1245 kbps)
 2009/07/01
 Chris@an
Timmerer,
Klagenfurt
University,
Austria
 12

  • 13. Demo
Video
 2009/07/01
 Chris@an
Timmerer,
Klagenfurt
University,
Austria
 13

  • 14. Conclusions
 •  Proxy
approach
for
in‐network
adapta@on
on
a
per‐packet
 basis
 –  Aware
of
sessions
and
individual
streams
(media‐aware)
 –  Stateful,
not
a
simple
packet
dropper
 •  Applica@ons
 –  Adapta@on
according
to
device
capabili@es
 –  Cross‐layer
adapta@on
 –  NAT
traversal
 •  Performance
sufficient
for
typical
home
deployments
 –  4
parallel
streams
with
30
percent
CPU
load
and
<
100
ms
delay
 
➙SVC
adapta@on
can
be
done
on
exis@ng
off‐the‐shelf
 network
devices
 2009/07/01
 Chris@an
Timmerer,
Klagenfurt
University,
Austria
 14

  • 15. References
 •  I.
Kofler,
J.
Seidl,
C.
Timmerer,
H.
Hellwagner,
I.
Djama
and
T.
Ahmed,
 “Using
MPEG‐21
for
cross‐layer
mul@media
content
adapta@on”,
Journal
 on
Signal,
Image
and
Video
Processing,
Springer,
vol.
2,
no.
4,
Dec.
2008.
 •  R.
Kuschnig,
I.
Kofler,
M.
Ransburg,
H.
Hellwagner,
“Design
op@ons
and
 comparison
of
in‐network
H.264/SVC
adapta@on”,
Journal
of
Visual
 Communica@on
and
Image
Representa@on,
Sept.
2008.
 •  I.
Kofler,
M.
Prangl,
R.
Kuschnig,
and
H.
Hellwagner,
“An
H.264/SVC‐based
 adapta@on
proxy
on
a
WiFi
router”,
Proceedings
of
the
18th
Interna@onal
 Workshop
on
Network
and
Opera@ng
Systems
Support
for
Digital
Audio
 and
Video
(NOSSDAV
2008),
Braunschweig,
Germany,
May
2008.
 •  I.
Kofler,
C.
Timmerer,
H.
Hellwagner,
and
T.
Ahmed:
“Towards
MPEG‐21‐ based
Cross‐layer
Mul@media
Content
Adapta@on”,
Proc.
2nd
 Interna@onal
Workshop
on
Seman@c
Media
Adapta@on
and
 Personaliza@on
(SMAP
2007),
London,
UK,
Dec.
2007.
 2009/07/01
 Chris@an
Timmerer,
Klagenfurt
University,
Austria
 15

  • 16. Thank
you
for
your
a;en@on
 ...
ques@ons,
comments,
etc.
are
welcome
…
 
Ass.‐Prof.
Dipl.‐Ing.
Dr.
Chris@an
Timmerer
 Klagenfurt
University,
Department
of
Informa@on
Technology
(ITEC)
 Universitätsstrasse
65‐67,
A‐9020
Klagenfurt,
AUSTRIA
 chris@an.@[email protected]‐klu.ac.at
 h;p://[email protected]/
 Tel:
+43/463/2700
3621
Fax:
+43/463/2700
3699
 ©
Copyright:
Chris.an
Timmerer
 2009/07/01
 Chris@an
Timmerer,
Klagenfurt
University,
Austria
 16