Coupled multi-agent simulations for mobile security & privacy research

Abstract

The complexity of mobile device digital ecosystems is rapidly increasing. Not only the absolute number of mobile devices, but also their diversity is on an ongoing rise. Devices are equipped with multiple short-range communication technologies that are used by different applications, which can become entry points for security or privacy threats. These threats need to be addressed in an early stage, while the mobile ecosystem is still relatively small and flexible. In prior work we presented the advantages of simulation approaches in the field of security and privacy research within the mobile domain. In this paper, we extend the Mobile Security & Privacy Simulator, which allows us to model the ecosystem of mobile devices and its surrounding environment with more details to achieve more realistic results. For this purpose we build up additional sub-simulations for discrete real world sub-scenarios like indoor places and connect them with the main map-based simulation. By this coupling we are able to better fit a simulation to the real world using tailored models for agents as well as for their environment depending on each sub-scenario's characteristics.