BDSsim: A Mesa-Based Simulator for Botnet Defense System

This paper proposes a new simulator, BDSsim, for Botnet Defense System (BDS). BDS is a cybersecurity system that uses white-hat botnets to exterminate malicious botnets. Conventionally, an agent-oriented Petri net,
$\text{PN}^{2}$
, has been used to study BDS. However, when trying to precisely model the behavior of BDS, the resulting net becomes too large, and some parts of the modeling had to be simplified. BDSsim describes BDS behavior in Python code and utilizes Mesa, an agent-based modeling and simulation framework in Python. BDSsim enables sophisticated modeling of BDS behavior and effective simulation utilizing Mesa and the Python ecosystem. The simulation with BDSsim results quantitatively revealed that (i) the primary infection rate of white-hat worms affects the effectiveness of BDS as well as the secondary infection rate and lifespan; (ii) the network topology, especially network density, affects the effectiveness of BDS.