National Technical University of Athens (NTUA)
Μελέτη της επίδρασης δειγματοληψίας στη διαδικασία ανίχνευσης ανωμαλιών στο διαδίκτυο
Abstract
dc:descriptionIn this work we study and evaluate the effect of sampling on the field of network anomaly detection in the Internet. Most of the existing work on sampling is related to general network management functions and focuses on the study of some general statistical traffic properties such as flow size distribution, average flow size, or total number of flows in a sample of network data. In the present thesis a novel sampling methodology suitable for application on the field of network anomaly detection is proposed and analyzed. The idea of the proposed sampling method is based on the observation that most of the network anomalies are caused by small flows (with a small number of packets). The proposed method is named “Selective Sampling” and it focuses on the dynamic and preferential selection of small flows (in packets) which are usually the source of many network attacks such as Distributed Denial of Service (DDoS) Attacks and worm propagation. In addition, we propose and analyze a method of two-stage sampling, combining suitably the sampling of flows and packets, while making use of the proposed “Selective Sampling” method. The results show that the proposed approach improves significantly the anomaly detection effectiveness, while at the same time decreases the number of selected data. Moreover, we study the effect of “Intelligent Sampling Methods” in the detection and classification of various anomalies on the network. We show that through sampling techniques that opportunistically and preferentially sample traffic data we achieve to “magnify” the appearance of anomalies within the sampled data set and therefore improve their detection. Therefore, the inherently “lossy” sampling process is transformed to an advantageous feature in the anomaly detection case, allowing the revealing of anomalies that would be otherwise untraceable and thus becoming the vehicle for efficient anomaly detection and classification. The evaluation of the proposed sampling methods is achieved through the use of different anomaly detection techniques on real network data that have been collected from the link that connects the National Technical University of Athens (NTUA) with the Greek Research and Technology Network (GRNET).
Degree
thesis:*- Grantor dc:publisher
- National Technical University of Athens (NTUA)
- Year dc:date
- 2009
Author and committee
dc:creator, dc:contributor.*- Authors dc:creator
-
- Androulidakis, Georgios
- Ανδρουλιδάκης, Γεώργιος
Subjects
dc:subject × 12- Ανίχνευση ανωμαλιών δικτύου
- Δειγματοληψία
- Δικτυακές μετρήσεις
- Ασφάλεια δικτύων
- Network anomaly detection
- Sampling
- Network measurements
- Network security
- Επιστήμες Μηχανικού και Τεχνολογία
- Επιστήμη Ηλεκτρολόγου Μηχανικού, Ηλεκτρονικού Μηχανικού, Μηχανικού Η/Υ
- Engineering and Technology
- Electrical Engineering, Electronic Engineering, Information Engineering
Rights
- Language dc:language
- gre
Identifiers
dc:identifier.*- Identifier
- 10.12681/eadd/18243