National and Kapodistrian University of Athens
Management of network and energy resources in cognitive and self-organizing wireless networks
Abstract
dc:descriptionThe efficient usage of network resources and the reduction of the consumed energy in modern communication systems that operate in a dense urban environment are challenging tasks, due to the complexity and the spatio-temporal dynamics of wireless networks. Self-organization is considered as one of the most promising paradigms for the management of networks that operate in highly dynamic and dense environments. In this thesis a novel approach is proposed to dynamically control the size and configuration of a wireless network for the effective utilization of network resources and energy saving. Firstly, we define the functional and software architecture for a self-organizing network, based on the cognitive cycle concept. Central to our approach is the organization of access points into clusters to facilitate local management and coordination of the deployed cognitive managers. In each cluster, the elected cluster head monitors usage of resources as well as the energy consumption changes during the transmission and reception, at both the access point and user equipment sides, and decides on the appropriate adaptation action. Specifically, a scheme for access point dynamic deactivation and reactivation is proposed for efficient use of radio resources and reduction of access point energy consumption. Also, re-allocation of selected frequency channels and formation of multi-hop relays among user equipment is used to diminish the effect of access point deactivation on user equipment energy consumption. The energy consumption improvement attained under the proposed scheme, at both the access point and the user equipment sides, is evaluated via simulation. In addition a scheme has been proposed and evaluated for the coordination of different self-optimization actions, which are concurrently triggered by the cognitive managers in a self-organizing network (coverage and capacity optimization, energy saving, interference). The goal is to resolve conflicts on configuration actions and dependencies among performance metrics, ensuring network stability. Finally, an algorithmic framework is proposed to enhance the effectiveness of cognitive managers. Specifically, the accumulated history of previous events and a feedback-based scheme are introduced for the evolution of the decision making algorithms that are used for the improvement of VoIP QoS in a congested WiMAX network. The proposed algorithmic framework has been evaluated using the FIRE Panlab experimental facilities.
Degree
thesis:*- Grantor dc:publisher
- National and Kapodistrian University of Athens
- Year dc:date
- 2013
Author and committee
dc:creator, dc:contributor.*- Authors dc:creator
-
- Kousaridas, Apostolos
- Κουσαρίδας, Απόστολος
Subjects
dc:subject × 16- Διαχείριση δικτύων
- Ασύρματα δίκτυα
- Διαχείριση πόρων
- Εξοικονόμηση ενέργειας
- Αυτο-οργάνωση
- Γνωσιακή επιστήμη
- Network management
- Wireless networks
- Resource management
- Energy saving
- Self-organization
- Cognition
- Φυσικές Επιστήμες
- Επιστήμη Ηλεκτρονικών Υπολογιστών και Πληροφορική
- Natural Sciences
- Computer and Information Sciences
Rights
- Language dc:language
- eng
Identifiers
dc:identifier.*- Identifier
- 10.12681/eadd/29169