Spatially Aware Malware Infection Modeling Framework

The existing body of research on malware simulation does not make full use of the topological and geographic location of the simulated malware-infected computers on the internet. We address this issue for creating a topologically-aware framework for modeling the spread of malware. Our framework can accommodate a variety of infection models, as well as different host capabilities and on/off patterns. The output of the simulation is a plausible scenario of how an infection could spread over time, at the individual IP-address granularity. The results can be used, for example, to generate semi-realistic malware traffic as input to further simulations and testing. We validate the results by comparing to real-life infection data from a malware network.