|
Title:
|
FUNCTION APPROXIMATION MODELS FOR NONINTRUSIVE PREDICTION OF VOIP QUALITY |
|
Author(s):
|
Mousa Al-akhras , Khalil El Hindi |
|
ISBN:
|
978-972-8924-86-7 |
|
Editors:
|
Hans Weghorn, Jörg Roth and Pedro Isaías |
|
Year:
|
2009 |
|
Edition:
|
Single |
|
Keywords:
|
Regression Trees, Model Trees, Quality of Service, E-Model, Non-Intrusive Voice Quality, VoIP. |
|
Type:
|
Full Paper |
|
First Page:
|
123 |
|
Last Page:
|
131 |
|
Language:
|
English |
|
Cover:
|
|
|
Full Contents:
|
click to dowload 
|
|
Paper Abstract:
|
Multimedia applications such as Voice over IP (VoIP) are one of the most important applications and traffic types in
todays networks. Almost all network management solutions and hardware appliances test for VoIP traffic for the
purposes of network management and packet shaping in addition to Quality of Service (QoS) provisioning as the quality
of the multimedia service should be acceptable to the user. To this end, a specialised solution is needed for the purpose of
measuring the multimedia and mainly the voice quality while the network is running, therefore an automated, nonintrusive
and an accurate measurement of quality should be in place. The E-Model proposed by the International
Telecommunication Union Telecommunication Standardisation Sector (ITU-T) achieves the above requirements, but it
needs the timely consuming subjective tests to calibrate some of its parameters. In this paper we propose the use of
regression and model tree methods to extend the E-model while avoiding the need for the subjective tests. The proposed
solutions are compared against the E-Model to prove their effectiveness. |
|
|
|
|
|
|
