Identification of Polymorphic Malware using Fractal Complexity Analysis

In the proposed collaborative research, the focus will be on the application of fractal complexity analysis in anomaly detection. Many features are hidden deep within time series information such as network traffic, and complexity analysis will facilitate the extraction of such features. Complexity analysis takes advantage of the self-similar structure which is found widely in nature and which has been shown to be evident in network traffic. These features will be used for the presence of Malware which has compromised a host system. The features will also be used to detect incoming denial of service attacks or other resource crippling behavior which would indicate intrusion or the attempted disruption of normal operations. This research will improve the ongoing research aims of the intern by including complexity analysis as a tool towards the detection of obfuscated forms of Malware. This work will also produce a deliverable in the form of a manuscript which will summarize the work carried out and the core findings.

Faculty Supervisor:

Ken Ferens

Student:

Partner:

National Institute of Informatics

Discipline:

Engineering

Sector:

Education

University:

University of Manitoba

Program: