- author = "Muhammad Adil Raja",
- title = "Real-time Non-intrusive Speech Quality Estimation of Voice over Internet Protocol using Genetic Programming",
- school = "Electronic and Computer Engineering, University of Limerick",
- year = "2008",
- address = "Ireland",
- month = jun,
- keywords = "genetic algorithms, genetic programming",
-
URL = "
http://www.cs.ucl.ac.uk/staff/W.Langdon/ftp/papers/adil-raja-thesis-2008.pdf",
- size = "165 pages",
-
abstract = "Telecommunications technologies are evolving at a
rapid pace. The old Public Switched Telephone Network
(PSTN) is being replaced with wireless and voice over
IP (VoIP) systems. This requires the service providers
to offer their services on competitive prices, on one
hand, and to ensure the interoperability of their
services over heterogeneous networks on the other.
Added to this is the challenge of keeping up with the
expectations of the clientele as regards quality of
service (QoS). Thus to enable the successful deployment
and functioning of a telecommunications network, it is
equally important to estimate the speech quality as it
may be perceived by the humans. Speech quality is a
subjective opinion, based on the human users experience
of a call. Recently, objective speech quality
assessment has become a very active research area. This
is an attempt to circumvent the limitations of
subjective analysis by simulating the opinions of human
testers algorithmically. There are two distinct
approaches to objective testing namely, intrusive and
non-intrusive. While intrusive techniques employ a
reference speech signal to estimate the quality of a
degraded one, the non-intrusive models do not enjoy
this privilege as they rely solely on features of the
signal under test.
The goal of this research was to derive superior non-intrusive speech quality estimation models. Model superiority was sought in a multi-objective sense: 1) enhancement of prediction accuracy of the derived models as compared to the previous ones. 2) model simplicity or parsimony was desired as it may enhance the computational efficiency. In this research this is achieved by employing a novel approach based on Genetic Programming (GP).
GP is a machine learning algorithm which coarsely emulates concepts adopted from natural evolution to automatically generate computer programs. Evolution is performed in the hope of finding a program or a symbolic expression that appropriately solves the problem under consideration. This potential benefit of benefit of GP has been used in this thesis.",
- notes = "Available from the University of Limerick library. See also \cite{Raja:book}",
