Συντάχθηκε 06-10-2014 11:15
από Esthir Gelasaki
Email συντάκτη: egelasaki<στο>tuc.gr
Ενημερώθηκε:
-
Ιδιότητα: υπάλληλος.
ΠΟΛΥΤΕΧΝΕΙΟ ΚΡΗΤΗΣ
Σχολή Ηλεκτρονικών Μηχανικών και Μηχανικών Υπολογιστών
Πρόγραμμα Μεταπτυχιακών Σπουδών
ΠΑΡΟΥΣΙΑΣΗ ΜΕΤΑΠΤΥΧΙΑΚΗΣ ΕΡΓΑΣΙΑΣ
ΔΗΜΗΤΡΙΟΥ ΚΩΝΣΤΑΝΤΙΝΑ
με θέμα
Revenue and Irritation-Based Call Admission Control Using Road Maps over Wireless Cellular Networks.
Τετάρτη 8 Οκτωβρίου 2014, 12πμ
Αίθουσα 145.Π42, Κτίριο Επιστημών, Πολυτεχνειούπολη
Εξεταστική Επιτροπή
Αναπληρωτής Καθηγητής Πολυχρόνης Κουτσάκης (επιβλέπων)
Καθηγητής Μιχαήλ Πατεράκης
Αναπληρωτής Καθηγητής Άγγελος Μπλέτσας
Abstract
Bandwidth remains the most valuable commodity in wireless networks, even more so today, as new bandwidth-hungry multimedia applications emerge constantly and their users have high Quality of Service (QoS) requirements. Hence, the ability of a wireless network to efficiently allocate its bandwidth resources is crucial.
In this thesis, we propose a new mechanism which controls the admission of users moving from cell to cell in a wireless cellular network. The Call Admission Control (CAC) mechanism is defined as a sequence of activities that are realized from the network in order to check if a user’s request, to use a specific service of the network, can be admitted or not. This request will be admitted if the desirable level of QoS can be accomplished without causing violation on the QoS that existing users enjoy. The design and simulation of the CAC mechanism were realized for the simultaneous use of the channel both from voice and e-mail users and video users downloading movies encoded with MPEG-4. The CAC mechanism aims to maximize channel throughput without allowing network congestion which would lead to the violation of the users’ QoS requirements. This is difficult to achieve because of the contradictory nature of the different types of traffic. Our work is one of the first in the relevant literature, to the best of our knowledge, to simultaneously consider both the provider revenue and user irritation in making its admission decisions.
Our results show the resource allocation efficiency of the proposed mechanism in two different implementations regarding the degradation/upgrades of video users. Our CAC mechanism is shown to provide voice and video users long term satisfaction with the QoS they receive, even for very high traffic loads, while at the same time allowing the provider to increase its profit.
