Detection of Malcodes by Packet Classification

Authors:
  • IAIrfan Ahmed
    Irfan Ahmed
  • KLKyung-Suk Lhee
    Kyung-Suk Lhee

Abstract
In this paper, we propose an anomaly detection approach that classifies packets into code-type and data-type. Our objective is to detect a packet containing codes flowing into a network port, which normally expects data packets only. The proposed approach can detect potentially malicious packets such as worms, viruses, and shellcodes. We propose a time-efficient algorithm and show the results of our initial experiments.