USING SCALABLE VIDEO CODING FOR
DYNAMIC ADAPTIVE STREAMING OVER HTTP IN MOBILE ENVIRONMENTS
Christopher Müller
1
, Daniele Renzi
2
, Stefan Lederer
1
, Stefano Battista
2
, and Christian Timmerer
1
1
Alpen-Adria-Universität Klagenfurt, Universitätsstraße 65-67, 9020 Klagenfurt, Austria, {firstname.lastname}@itec.aau.at
2
bSoft ltd, 156 via Velini, 62100 Macerata, Italy, {firstname.lastname}@bsoft.net
ABSTRACT
Dynamic Adaptive Streaming over HTTP (DASH) is a
convenient approach to transfer videos in an adaptive and
dynamic way to the user. As a consequence, this system
provides high bandwidth flexibility and is especially
suitable for mobile use cases where the bandwidth variations
are tremendous. In this paper we have integrated the
Scalable Video Coding (SVC) extensions of the Advanced
Video Coding (AVC) standard into the recently ratified
MPEG-DASH standard. Furthermore, we have evaluated
our solution under restricted conditions using bandwidth
traces from mobile environments and compared it with an
improved version of our MPEG-DASH implementation
using AVC as well as major industry solutions.
Index Terms— Dynamic Adaptive Streaming over
HTTP, MPEG-DASH, Scalable Video Coding, Evaluation,
Mobile Networks, Vehicular Mobility
1 INTRODUCTION
Dynamic Adaptive Streaming over HTTP (DASH) has the
potential to play a major role in networks with fluctuating
bandwidth. The major industry players (e.g., Microsoft,
Apple, and Adobe) have already adopted it and several
streaming providers like Netflix, Hulu, Vudu, and Amazon,
which are using the Hypertext Transfer Protocol (HTTP) for
their streaming service. One major advantage of HTTP is its
convenience for the end user, as well as for the streaming
provider. Nevertheless, TCP and HTTP introduces a
significant overhead compared to RTP and UDP, which is
definitely a disadvantage [1]. Another fact that cannot be
handled by traditional HTTP streaming (i.e., progressive
download) are varying bandwidth conditions.
Dynamic Adaptive Streaming over HTTP aims to
address this fact with a quite simple yet effective approach.
Instead of having one media file encoded at a single bitrate,
the same media file will be encoded at several bitrates,
resolutions, etc. These multiple versions of same media will
be then chopped into segments that can be individually
requested by the client through HTTP. This enables the
client to switch between different qualities, resolutions, etc.
during the streaming session. Furthermore, the clients can be
served through ordinary Web servers, which let the system
scale very well. As mentioned at the beginning the industry
has already deployed several solutions and also MPEG has
recently ratified DASH as international standard [2].
Typically, AVC will be used to generate multiple
qualities of the media for DASH but also the scalable
extensions [3] of AVC are suitable and can potentially bring
some major advantages due to its layered architecture,
which enhances the flexibility of the segment selection. That
is, in comparison to AVC it is possible to cancel a segment
request at the layer boundaries. That advantage could
simplify the adaptation process because this allows the
client trying to download up to the highest quality and in
case of insufficient bandwidth it could cancel the request at
the segment boundaries. This is not possible with AVC
because when the client cancels a segment download, the
video data of that segment becomes useless.
The goal of this paper is to improve our existing MPEG-
DASH implementation using AVC [4] and evaluate a SVC-
based solution on top of it. In particular, we have compared
both solutions among themselves as well as with the major
industry solutions. In anticipation of the results we can
conclude that we have achieved a major improvement of our
own implementation [4], which performs now better than all
industry solutions. Furthermore, our SVC-based solution
could utilize a higher overall bandwidth compared to AVC.
The remainder of this paper is organized as follows.
Section 2 describes related work and Section 3 describes our
integration of SVC into MPEG-DASH. The experimental
setup and results are described and discussed in Section 4.
Finally, the paper is concluded in Section 5.
2 RELATED WORK
Sanchez et al. [5] has already described potential benefits of
SVC with DASH. However, their main focus is on cache
performance and encoding while we focus on the adaption
process itself within mobile environments. Kofler et al. [6]
has evaluated the implications of the ISO Base Media File
Format (ISOBMFF) on adaptive HTTP Streaming of SVC
and shown that the ISOBMFF is not suitable for SVC
streams below 1 Mbps. Akhsabi et al. [7] evaluated
Microsoft Smooth Streaming, Adobe HTTP Dynamic
Streaming, and the Netflix Player using simulated
20th European Signal Processing Conference (EUSIPCO 2012) Bucharest, Romania, August 27 - 31, 2012
© EURASIP, 2012 - ISSN 2076-1465 2208