Introduction

Motivation

In the age of the Internet of Things, embedded communicating systems are becoming massively widespread in critical infrastructures. They contribute to a better control and optimization of these infrastructures to increase their efficiency, cost and use, but also to meet societal challenges. Unfortunately, they also contribute to the increase of the global attack surface of information systems, which represents an unprecedented threat.

It is therefore essential to guarantee the best level of protection for such systems that handle sensitive or secret data. Indeed, because of their connectivity, they are subject to numerous software and hardware threats. In the context of the TrustGW project, the system under consideration is composed of objects connected to a gateway, which in turn is connected to one or more calculation servers. According to an ENISA report, 43% of processing related to the Internet of Things will be carried out at the gateway level by 2021. The connected objects (IoT nodes) transmit and receive data from the gateway. Each node potentially communicates with a different waveform (e.g. LoRaWan and Bluetooth). Indeed, the multiplication of waveforms limits the generalization of a single waveform for all communications within this type of infrastructure. The gateway must therefore be able to support different waveforms in a dynamic way.

The gateway architecture, which is at the heart of the project, is heterogeneous (software-hardware), composed of a baseband processor (BBP), an application processor (GPP) and hardware accelerators implemented on an FPGA. The latter are deployed dynamically according to the acceleration needs of the applications at a given time. The dynamic reconfiguration allows to specialize the gateway during execution. The FPGA resources are virtualized in order to have a uniform view from the applications’ point of view. This type of architecture makes it possible to reach the required performance levels while respecting the power constraints essential to the targeted domain. The gateway embeds several virtual machines in order to be able to deploy the services of the different execution domains that it hosts in a partitioned manner. These virtual machines share certain computing resources (processor and FPGA) and memory resources for reasons of infrastructure cost, maintenance and standardization of the architectures implemented. Nevertheless, some resources must be exclusive according to the services available within an execution domain. Hypervisors are used to deploy virtual machines and ensure their isolation.

Objectives

The TrustGW project therefore aims to develop a dynamically reconfigurable heterogeneous software-hardware gateway architecture that can be trusted. The implementation of such an architecture guaranteeing confidentiality, integrity, availability and authentication properties is original. In the framework of the TrustGW project, three main scientific challenges are addressed:
  • To design a heterogeneous software-hardware gateway architecture that is trusted and dynamically reconfigurable;
  • To propose a trusted hypervisor that allows the deployment of virtual machines on a heterogeneous software-hardware architecture with resource virtualization;
  • Guarantee the security of applications within virtual machines.