A system approach to network modeling for DDoS detection using a Na
Authors:
- 1 January 2011
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- No. 21552487,p. 1-10
Abstract
Denial of Service(DoS) attacks pose a big threat to any electronic society. DoS and DDoS attacks are catastrophic particularly when applied to highly sensitive targets like Critical Information Infrastructure. While research literature has focussed on using various fundamental classifier models for detecting attacks, the common trend observed in literature is to classify DoS attacks into the broad class of intrusions, which makes proposed solutions to this class of attacks unrealistic in practical terms. In this work, the approach to a carefully engineered, practically realised system to detect DoS attacks using a Naìve Bayesian(NB) classifier is described. The work includes network modeling for two protocols - TCP and UDP.Keywords
This publication has 11 references indexed in Scilit:
- From Feature Selection to Building of Bayesian Classifiers: A Network Intrusion Detection PerspectiveAmerican Journal of Applied Sciences, 2009
- Defending DDoS Attacks Using Hidden Markov Models and Cooperative Reinforcement Learning2007
- A Novel Model for Detecting Application Layer DDoS Attacks2006
- A machine learning framework for network anomaly detection using SVM and GA2005
- Real-time detection of distributed denial-of-service attacks using RBF networks and statistical featuresComputer Networks, 2005
- Mark-aided distributed filtering by using neural network for DDoS defense2005
- A taxonomy of DDoS attack and DDoS defense mechanismsACM SIGCOMM Computer Communication Review, 2004
- Statistical approaches to DDoS attack detection and response2004
- DDoS detection based on feature space modeling2004
- Learning nonstationary models of normal network traffic for detecting novel attacks2002