Section: Application Domains

Cyber-Physical Systems

Cyber-Physical Systems (CPSs) used to be well isolated and so designed accordingly. In the last decade, they have become integrated within larger systems and so accessible through the Internet. This is the case with industrial systems, like SCADA, that have been unfortunately exposed to major threats. Furthermore, the Internet-of-Things (IoT) has become a reality with numerous protocols, platforms and devices being developed and used to support the growing deployment of smart* services: smart home, transport, health, city... and even rather usual rigid systems such as industry 4.0.

From an academic perspective, the IoT can be seen as an evolution of sensor networks. It thus inherits from the same problems regarding security and scalability, but with a higher order of magnitude both in terms of number of devices and their capabilities, which can be exploited by attackers. Research in this area has focused on developing dedicated protocols or operating systems to guarentee security and performance, Resist aims to tackle identical problems but assuming a more practical deployment of IoT systems composed of heterogeneous and uncontrolled devices. Indeed, this ecosystem is very rich and cannot be controlled by a unique entity, e.g. services are often developed by third parties, manufacturers of embed devices are different from those providing connectivity.

As a result, managing an IoT system (monitoring, changing configuration, etc.) is very hard to achieve as most of the devices or applications cannot be directly controlled. For instance, many IoT providers rely on their own cloud services, with their own unknown proprietary protocols and most of the time through encrypted channels. Above all, the use of middle-boxes like gateways hides the IoT end-devices and applications. We will thus need to infer knowledge from indirect and partial observations. Likewise, control will be also indirect for example through filtering or altering communications.