AAAI Publications, The Twenty-Eighth International Flairs Conference

Font Size: 
Network Intrusion Detection Using a Hardware-Based Restricted Coulomb Energy Algorithm on a Cognitive Processor
Mahdi H. Moghaddam, Ricardo A. Calix

Last modified: 2015-04-07

Abstract


The current state of the art in intrusion detection systems mainly relies on heuristic rules called signatures to detect intrusions to a network environment. The downside of signature based approaches is that they can only detect previously known attacks. Since no signature exists for new attacks, other approaches need to be considered. Here, machine learning algorithms may be beneficial. Additionally, at the network level, intrusion detection system performance is very important. Therefore, fast and efficient machine learning implementations are needed. In this study, a parallel hardware based implementation of the KNN and RCE classifiers will be analyzed to get a better understanding of the advantages and disadvantages of hardware based machine learning for network intrusion detection.

Keywords


IDS; CM1K

Full Text: PDF