Aristotle University Of Thessaloniki (AUTH)
Αλγόριθμοι χρονοδρομολόγησης για ασύρματα συστήματα εκπομπής δεδομένων
Abstract
dc:descriptionData broadcasting has emerged as an efficient means for the dissemination of information over asymmetric wireless networks. Examples of data broadcasting applications are traffic information, weather information, news distribution systems, sightseeing applications. In such applications, client needs for data items are usually overlapping. Consequently, broadcasting seems to be an efficient solution, as the broadcast of a single information item will likely satisfy a (possibly large) number of client requests. In this thesis, four scheduling algorithms for wireless data broadcasting systems, and more specifically for wireless push systems, that are able to adapt their broadcast schedule to the characteristics of the environments where they operate, are proposed. The proposed algorithms achieve high performance and are able to adapt efficiently to the demands of the operational environment. Firstly, a wireless push system for environments with unequal client group sizes and different service priority levels is proposed. Taking into account the population and the priority level of each group, the broadcast server forms the broadcast schedule of the system so that fair service for all clients along with efficient service of the different priority levels of the system are achieved. In another system, the use of smart antennas at the broadcast server, instead of directional antennas with fixed beamwidth is proposed. The ability of smart antennas to change their beamwidth is exploited so that the coverage area of each antenna is changed according to the current placement of clients within the system. The simulation results reveal that the use of smart antennas instead of the fixed-beadmwidth antennas has led to a significant improvement of the system’s performance. Furthermore, a distributed adaptive push system is proposed. The novelty of the introduced system has to do with the replacement of the single centralized learning automaton of the broadcast server with a number of distributed learning automata that are equal to the number of clients’ groups. The proposed scheme seems to be more sensitive concerning the various client demands compared to the centralized one. The simulation results show that the proposed distributed scheme is more efficient and more fair compared to the centralized one. Finally, in this thesis, the use of Hidden Markov Models has been introduced for the first time as a prediction module of the client demands in a wireless push system. The proposed system uses a predictive module based on the Hidden Markov Model aided by a set of history vectors. The outcome of the Hidden Markov Models is used to conduct a schedule of the forthcoming server’s broadcasts. The simulation results reveal that the proposed scheme succeeds accurate predictions of the clients’ demands, resulting in an efficient operation of the system.
Degree
thesis:*- Grantor dc:publisher
- Aristotle University Of Thessaloniki (AUTH)
- Year dc:date
- 2012
Author and committee
dc:creator, dc:contributor.*- Authors dc:creator
-
- Kakali, Vassiliki
- Κακάλη, Βασιλική
Subjects
dc:subject × 20- Ασύρματα συστήματα ώθησης
- Δικαιοσύνη
- Επίπεδα προτεραιότητας
- Έξυπνες κεραίες
- Κατανεμημένα μανθάνοντα αυτόματα
- Κρυμμένο μαρκοβιανό μοντέλο
- Πρόβλεψη
- Προσαρμογή
- Wireless push systems
- Fairness
- Priority levels
- Smart antennas
- Distributed learning automata
- Hidden Markov Model
- Prediction
- Adaption
- Φυσικές Επιστήμες
- Επιστήμη Ηλεκτρονικών Υπολογιστών και Πληροφορική
- Natural Sciences
- Computer and Information Sciences
Rights
- Language dc:language
- gre
Identifiers
dc:identifier.*- Identifier
- 10.12681/eadd/27398