Tiny, PrivAte, Proved & isolaTed – TinyPART
In the next 10 years, SD-IoT (Software-Defined IoT) is expected to evolve enabling flexible programmability of IoT devices. In this paradigm, some part of the logic (e.g. pre-processing of data) will be transferred on demand from the cloud to the device for privacy reasons. Relying in particular on low-power devices based on microcontrollers, SD-IoT will control and finely monitor everything in our physical world in the future. In this context, enabling and securing privacy-by-design for IoT data on such devices becomes a crucial task. In particular, the current situation where all parts of the code & data (e.g. a 3rd party network stack, cryptography or personal data) of a constrained device share a common address space is infeasible and requires new tools and platforms.
The project TinyPART will leverage SD-IoT and design a deeply embedded software platform fitting low-power devices, architectured for privacy-by-design and cybersecurity. The key features of the TinyPART platform include mechanisms for memory isolation, containerization of code, so-called microContainers, as well as microPrivacy and lightweight cryptography, a toolset enabling privacy of IoT data. The platform is tailored to applications, where updatable snippets of 3rd party code (or possibly interpretable script, such as Python) are hosted on the IoT devices accessing, pre-processing and sharing protected local data. Notably, by contrast, current software on low-power devices is typically monolithic, and maintained by a single tenant -- which is not adequate.
TinyPART brings together complementary expertise from partners both in France and Germany with multiple previous joint collaborations (RIOT open source community, PIP User Club, Celtic-Plus ODSI project, BMBF SecureFOG and PROPHYLAXE, ....). Their expertise ranges from formal proof & software co-design, embedded systems operating systems design, remote provisioning & management, and privacy-preserving algorithms, particularly differential privacy and lightweight cryptography for IoT. Within their areas all partners have already made significant impact in the development of low power IoT.
TinyPART based on RIOT/PIP can hope to have a very important socioeconomic impact in the future, similar to the impact of Linux for instance, with indirect business models based on the free and transparent embedded software platform (open source) offered by RIOT.
Madame Chrystel Gaber (ORANGE SA)
The author of this summary is the project coordinator, who is responsible for the content of this summary. The ANR declines any responsibility as for its contents.
CRIStAL Centre de Recherche en Informatique, Signal et Automatique de Lille
ORANGE ORANGE SA
Inria Saclay-Ile de France -équipe TRIBE Centre de Recherche Inria Saclay - Île-de-France
Help of the ANR 399,555 euros
Beginning and duration of the scientific project: May 2021 - 36 Months