Abstract

The virtualization of services and network functions is one of the base technologies used massively in 5G networks. It allows designing and deploying new services in a simple and scalable way. But it also introduces the need to manage the security of network elements that are deployed in the cloud, in an infrastructure that is not, in most cases, owned by network providers. In this context, the concept of Level of Trust (LoT) arises as a metric that allows reviewing how reliable the virtualized elements of a new function or service are as a preliminary step to its deployment in a network infrastructure that, in turn, time, it will be virtualized. An intelligent management of the Level of Trust should allow the automatic verification of the LoT in the virtualized elements that are going to be deployed, thus accelerating the process of accepting these new services. An Attestation Software element will help verify the reliability of a platform and the integrity of the binaries running on it, in this case virtualized network functions and services. On the other hand, a tool will be developed to generate Topology Attestation that guarantees that the traffic flow follows the path that has been predetermined for a service, that it has passed through certain network elements and, furthermore, in the predetermined order. This is a concept that is being defined right now in the standards bodies so this would be a very early implementation. Finally, an Artificial Intelligence module will generate knowledge from the attestation data. This can be translated into corrective measures that allow the level of confidence to be improved in the successive executions of the services. The application context of the proposed system is that of advanced 5G (Beyond 5G or B5G) and, especially, that related to Network Slicing. This is a way to virtually separate vertical application domains that share similar quality of service and security needs and where virtual network segments (slices) typically use infrastructure owned by multiple operators.