Intelligent Detection of Cyberattacks in Novel Networks

Programmable networks refer to computer networks that can be dynamically configured and controlled through software rather than relying solely on static configurations. In simple terms, it means that the behavior and functionality of a network can be modified and customized using software-based programming, allowing for greater flexibility and adaptability. P4 (Programming Protocol-Independent Packet Processors) is a programming language used for defining how network packets are processed in programable devices, such as switches and routers. It is designed to be protocol-independent, meaning it is not tied to any specific networking protocol, and provides a flexible and programmable way to define how packets are processed and forwarded in a network. However, the fact of running code in programmable switches and routers opens the door to cyberattacks and misbehaviors. 

This thesis/project is proposed in collaboration with researchers from the Universidade Federal do Rio Grande do Sul, in Brazil, and presents the following objectives: i) create a programmable network (using real or virtualized programmable switches/routers) and inject traffic from an existing dataset, ii) execute a cyberattack in the switch network, and iii) monitor the internal behavior of the switch (CPU, RAM, etc.), and detect misbehaviors using a machine learning-based approach (or another one).

Bonati, L., Polese, M., D’Oro, S., Basagni, S., & Melodia, T. (2020). Open, programmable, and virtualized 5G networks: State-of-the-art and the road ahead. Computer Networks, 182, 107516.

Hauser, F., Häberle, M., Merling, D., Lindner, S., Gurevich, V., Zeiger, F., ... & Menth, M. (2023). A survey on data plane programming with p4: Fundamentals, advances, and applied research. Journal of Network and Computer Applications, 212, 103561.

Sánchez, P. M. S., Valero, J. M. J., Celdrán, A. H., Bovet, G., Pérez, M. G., & Pérez, G. M. (2021). A survey on device behavior fingerprinting: Data sources, techniques, application scenarios, and datasets. IEEE Communications Surveys & Tutorials, 23(2), 1048-1077.