National Technical University of Athens (NTUA)
Εξοικονόμηση ενέργειας σε αυτο-οργανούμενα δίκτυα ανεκτικά σε καθυστέρηση
Abstract
dc:descriptionThis thesis deals with energy-efficiency in wireless self-organizing delay tolerant net-works (DTNs) comprising autonomous, energy-constrained nodes. In particular, the thesis addresses energy consumption under three critical aspects of DTNs; the information overhead diffused in the network, the controlled mobility of DTN nodes and the active sensing they perform. The dynamic DTN nature, resulting from frequent disconnections of inter-nodal links, which is due to either the exhaustion of energy resources or to node sparsity, motivates the autonomous DTN nodes to perform either of the following operations (i) transmit multiple copies of stored information to ensure that the information reaches its destination (ii) move, in order to increase the encounter probability with nodes within their range or (iii) perform active sensing to exploit any chance of establishing communication with neighboring nodes. These operations may be extremely energy consuming if not properly designed. Hence, in the attempt to optimize specific network performance metrics, the relevant schemes should take into account the residual energy resources of DTN nodes. The thesis addresses the issue of information overhead diffusion in DTNs by proposing a scheme based on Network Coding (NC). As opposed to traditional routing/forwarding protocols employed in DTNs that store and forward replicas of received information, the NC technique allows the intermediate nodes to process the received packets and forward a combination of them to the next node on the end-to-end path; when linear NC is employed, the packets are linearly combined before transmission. The thesis applies the random linear NC (RLNC), which is suitable for self-organized networks with autonomous nodes. This is accomplished by multiplying the stored information with linear coefficients randomly selected by the DTN nodes from a sufficiently large finite field. The proposed NC/MMA scheme implements NC both for storing and forwarding the received information to the next node on the end-to-end path. Extensive simulations have shown that the proposed NC/MMA scheme achieves a drastic reduction (i) of the information overhead diffused in the DTN and (ii) of the energy consumption due to transmission, thus extending the network lifetime. In the framework of controlled mobility, the thesis proposes a DTN mobility scheme that regulates the nodes movement based on their residual energy resources and the contact rate characterizing the network. It has been verified in the literature that mobility increases the nodes encountering probability. However, the nodes mobility induces an energy cost due both to the nodes physical movement and to the increased sensing frequency. This energy mobility cost gives rise to the following dilemma; should DTN nodes move in order to increase their encounter probability or should they stay static expecting that other nodes will move? This dilemma along with the respective trade-off between mobility cost and encounter probability are dealt with by the thesis. To formulate the relevant DTN mobility problem, game theory is employed as it can effectively deal with such trade-offs. Extensive simulations have shown that the proposed scheme is capable of properly handling the trade-off between energy consumption and encounter probability, offering a significant reduction in energy consumption compared to standard mobility models, at the cost of a considerably smaller reduction in encounter probability. In wireless self-organizing DTNs amenable to encounter probability reduction, the proposed scheme, being capable of regulating nodes mobility, may offer a significant lifetime extension. The thesis also addresses the subject of active sensing in DTNs. An autonomous scheme is proposed that regulates nodes active sensing, taking into account their residual energy resources. Although contact probing is the only way of establishing communication between neighboring nodes, when not properly designed, it may lead to unnecessary energy consumption or to missed chances of establishing communication between nodes in sparse DTNs. The energy cost induced due to network probing gives rise to the following dilemma; should DTN nodes perform contact probing in order to increase the probability of establishing connection with neighboring nodes or not, expecting that other nodes will perform probing? This dilemma along with the respective trade-off between the energy cost due to probing and the connection probability are dealt with by the thesis employing game theory. Extensive simulations have shown that the proposed scheme handles the trade-off between energy consumption for probing and connection probability, offering a significant reduction in energy consumption, at the same time preserving a minimum acceptable connection probability level. In wireless self-organizing DTNs amenable to connection probability reduction, the proposed scheme, being capable of regulating nodes active sensing, may offer a significant lifetime prolongation. Concluding, the thesis contributes in the area of energy-efficiency in wireless self-organizing delay tolerant networks.
Degree
thesis:*- Grantor dc:publisher
- National Technical University of Athens (NTUA)
- Year dc:date
- 2015
Author and committee
dc:creator, dc:contributor.*- Authors dc:creator
-
- Vazintari, Anna
- Βαζιντάρη, Άννα
Subjects
dc:subject × 19- Αυτο-οργανούμενα δίκτυα ανεκτικά σε καθυστέρηση
- Κωδικοποίηση δικτύου
- Ελεγχόμενη κινητικότητα
- Ενεργητική ανίχνευση
- Θεωρία παιγνίων
- Delay tolerant networks
- Wireless self-organized networks
- Network coding
- Controlled mobility
- Contact probing
- Game theory
- Φυσικές Επιστήμες
- Επιστήμη Ηλεκτρονικών Υπολογιστών και Πληροφορική
- Επιστήμες Μηχανικού και Τεχνολογία
- Επιστήμη Ηλεκτρολόγου Μηχανικού, Ηλεκτρονικού Μηχανικού, Μηχανικού Η/Υ
- Natural Sciences
- Computer and Information Sciences
- Engineering and Technology
- Electrical Engineering, Electronic Engineering, Information Engineering
Rights
dc:rights- Statement dc:rights
-
- BY_NC_ND
- Language dc:language
- gre
Identifiers
dc:identifier.*- Identifier
- 10.12681/eadd/43337