부산대학교 도서관

5G and subsequent cellular network generations aim to extend ubiquitous connectivity of billions of Internet-of-Things (IoT) for their consumers. Security is a prime concern in this context as adversaries have evolved to become smart and often employ new attack strategies. Network defenses can be enhanced against attacks by employing behavior models for devices to detect misbehavior. One example is abusive modeling (AM) that is inspired by financial technologies to defend adversaries operating with unlimited resources who have no intention of self-profit apart from harming the system. This article investigates behavior modeling against abusive adversaries in the context of 5G and beyond security functions for IoT. Security threats and countermeasures are discussed to understand AM. A complexity–security tradeoff enables a better understanding of the limitations of state-based behavior modeling and paves the way as a future direction for developing more robust solutions against AM.