Dependable networks

Introduction

Our digital society heavily depends on the correct operation of networks and the systems connected to these networks. There are two kinds kinds of networks. The first kind are core networks, including the Internet, local networks (WIFI, Zigbee) and mobile networks (4/5G). The second kind perform a dedicated task, together with their connected systems and applications. This second kind of networks are sometimes called cyber physical systems, or embedded systems. Examples include SCADA (Supervisory Control And Data Acquisition) systems, energy systems and networks for Vehicle to Vehicle (V2V) and Vehicle to Infrastructure (V2I) communications.

A key research challenge that students of this track will be confronted with, is the question of how to improve the dependability of these networks.

Core networks, such as the Internet, are confronted with all kinds of security and privacy threats. Attacks on core Internet services, such as DNS and BGP, can have a severe effect on the operation of the Internet. DDoS attacks have the potential to stop crucial services and disturb our society. Errors in Internet of Things (IoT) devices can not only be misused for (botnet) attacks, but also to steal personal data and break privacy. Backdoors installed by some nation states in core switches and routers undermine our digital sovereignty. For this kind of security research, measurements are generally a first step, to better understand the details of the security and privacy threats. Once the problem is understood, mitigation actions can be proposed and implemented.

As opposed to the Internet, the concept of cyber physical systems is relatively new and such systems are sometimes still in early phases of the design. Networks for vehicular communications, for example, are under active development by organisations such as IEEE. Research challenges include the modelling and design of geographical routing, to ensure that certain messages (like traffic congestion) will only be received by cars on a certain highway, and not by cars on neighbouring streets. In the case of SCADA systems (systems that control critical infrastructures, such as power plants and energy distribution networks) it is important to make models that predict the effect of possible cyber attacks, to make to system more resilient.