Liebe Kolleginnen und Kollegen, liebe Studierende,
am Montag, dem 12.05.2014 um 17:15 h spricht Herr Kollege Hellwagner von
der Universität Klagenfurt in unserem Fakultätskolloquium in
Raum A 5,6 C 012 über das Thema
In-Network Adaptation for Adaptive Digital Video Streaming
Hier das Abstract:
Modern digital video portals and streaming services typically store
multiple, differently encoded variants of the available videos and
employ adaptation techniques in order to adapt the volume and quality of
the transmitted video data, in response to, e.g., dynamically varying
network throughput conditions or specific constraints of end devices.
Traditionally, adaptation has been performed by the video server.
Recently, though, client-driven adaptive delivery techniques have been
gaining importance in practice, e.g., the standardized Dynamic Adaptive
Streaming over HTTP (DASH) scheme or its proprietary variants. Besides
DASH-type schemes, the Multimedia Communication research group has for
several years explored yet another option, namely video adaptation in
the network proper, i.e., on network devices, so-called Media-Aware
Network Elements (MANEs). The talk will present experiments, results,
and potential applications of this In-Network Adaptation (INA) option.
We will first describe the basic design considerations, mechanisms,
prototype realizations, and performance results of scalable video
content (H.264/SVC) adaptation on MANEs. We have realized SVC adaptation
proxies on inexpensive, Linux-based WiFi home routers and demonstrated
that a reasonable number of SD SVC streams can be transmitted and
adapted simultaneously. The benefits of in-network adaptation on these
home routers will be illustrated by streaming HD SVC content over an
IEEE 802.11 wireless network and by timely adaptation of the video
according to the varying conditions on the wireless link toward a mobile
client.
Then we will discuss how such video adaptation mechanisms can be made
use of in the context of a swarm of Unmanned Aerial Vehicles (UAVs)
which explore an area, e.g., in a disaster response operation to find
missing or injured persons, and transmit videos to a base station, if at
all possible. In case of deteriorating network conditions, e.g., due to
increasing distance from the base station, a UAV capturing a video
(H.264/AVC) may autonomously reduce the video quality and data volume in
order to continue delivering useful video material. The system
architecture and the basis for autonomous adaptation decisions will be
addressed.
Alle sind herzlich dazu eingeladen.
Beste Grüße,
Wolfgang Effelsberg
--
Prof. Dr. Wolfgang Effelsberg Tel: +49 621 181 2600
Praktische Informatik IV Fax: +49 621 181 2601
University of Mannheim effelsberg(a)informatik.uni-mannheim.de
68131 Mannheim, Germany pi4.informatik.uni-mannheim.de/~effels