The Application of the Graph Laplacian in Network Forensics

Abstract

Understanding the behavior of a virus spreading inside a computer network is crucial. In recent years, devices have become increasingly interconnected via communication protocols. Since computer networks are more ubiquitous than ever before, one infected device threatens several others. In order to better defend against network infections, the current methods need more novel means to combat network infections. Spectral graph theory may assist the study of network forensics by better understanding how a virus spreads throughout a networked system. This paper analyzes the spread of a virus throughout a computer network using actual data. The novelty is the calculation of various Laplacian matrices to observe the changes to the network, modeled as a graph resulting from the virus spread. Our discoveries may provide implications to gain early awareness and insights into the infectious situation.