Implementing a broadcast storm attack on a mission-critical wireless sensor network

Abstract

In this work, we emphasize the practical importance of mission-critical wireless sensor networks (WSNs) for structural health monitoring of industrial constructions. Due to its isolated and ad hoc nature, this type of WSN deployments is susceptible to a variety of malicious attacks that may disrupt the underlying crucial systems. Along these lines, we review and implement one such attack, named a broadcast storm, where an attacker is attempting to flood the network by sending numerous broadcast packets. Accordingly, we assemble a live prototype of said scenario with real-world WSN equipment, as well as measure the key operational parameters of the WSN under attack, including packet transmission delays and the corresponding loss ratios.We further develop a simple supportive mathematical model based on widely-adopted methods of queuing theory. It allows for accurate performance assessment as well as for predicting the expected system performance, which has been verified with statistical methods.