Présentation

The research activities of the “Network and Security” group focus on optimized and secured wireless communication protocols.

1 - The research activities on networking focus on the transmission of data, while providing the quality of service required by applications. Several types of wireless networks are studied: low-power wireless personal area networks (IEEE 802.15.4, ZigBee), low power wide area networks (LoRaWAN, Sigfox), cellular networks, Wi-Fi networks, ad hoc networks, mobile networks, vehicular networks (VANET), body area networks, etc. The typical targeted applications are monitoring applications in constrained environments (forests, fields, volcanoes, mines, airplanes, smart cities, etc.) or in e-health. The quality of service requirements include delay, throughput, energy, as well as security (confidentiality, integrity, authentication). Our contributions mainly focus on the design, the analysis, the simulation or the experimentation of innovative wireless network protocols, based on the medium access layer or on the network layer. Here are a few examples of the topics we study:

• Industrial monitoring: We study the MAC/NWK protocols for networks operating in confined or hostile areas, and having important requirements in terms of quality of service (typically delay and throughput). The targeted applications are the monitoring of production chains, of nuclear power plants, of submarines, of airplanes, etc.
• Environmental monitoring: We study the MAC/NWK protocols for networks that are deployed outside, with severe energy constraints. These networks often have to scale in the number of nodes. The targeted applications are the monitoring of volcanoes or mountains, of forests, or of fields.
• Networks to help people: Wireless networks are part of several recent technologies that enable to help elderly people or people with reduced mobility. Here, we consider the networks globally, from the integration of a sensor in an object used daily by the patient, to the user experience. The targeted applications include wheelchairs for people with reduced mobility, or walking canes for people with reduced vision.
• Vehicular networks: We aim at simulating and prototyping MAC/NWK protocols for vehicular ad hoc networks (VANET), in intelligent transport systems applications.

2 - The research activities on security are structured around the following topics:

• Security for the 5G: We study how to secure communication protocols for the 5G. For instance, we study how to establish session keys between connected objects. For instance, in the ANR MOBIS5 project, we study the security of the secured messaging application SIGNAL.
• Security of symmetric ciphers: In the ANR DECRYPT project, we analyze the security of symmetric ciphers. Our objective is to use artificial intelligence tools, such as constraint programming, in order to automatically cryptanalyze these ciphers.
• Security of applications: With the growth of Internet, the number of security protocols is increasing. We believe it is crucial to analyze the security of these cryptographic protocols, and to define the security properties of the new use cases. For instance, in the ANR Severitas project, we analyze the security of e-exams. We also study the security of e-voting protocols, e-auctions protocols, etc.
• Biometric security: The ANR Privabio project targets the secure authentication and identification of persons. Our objective is to study the security of such protocols, and to propose innovative approaches to improve their security.

In the following, we give a few other examples of our research activities:

• We propose MAC/NWK protocols for wireless networks intended to improve the performance of IEEE 802.15.4. For instance, we have designed a protocol that can handle two types of quality of service: an unconstrained traffic, transmitted in time intervals with contention, and a priority traffic, which can be forwarded across the whole network in a bounded number of time periods. Managing these two types of traffic causes issues in the dimensioning of the time slots. Moreover, if the MAC layer uses time available in one time slot in order to transmit frames from the other time slot, complex routing issues may arise (indeed, incoherent routing decisions might be applied on the same packet, as the packet follows rules from two potentially different routing protocols). We have studied how to avoid routing loops when the routing protocol changes from one hop to another, based on the decisions of the MAC layer.
• We study how to decode frames in collisions in the LoRa (long range) protocol, in order to increase the limited throughput of this protocol. Our algorithms are based on temporal information coming from a small (but controlled) desynchronization between the transmitters. The results we obtained through simulation and experiments show that we can decode several frames in collision.
• We have studied the impact of interferences between Wi-Fi and IEEE 802.15.4. This study enabled us to evaluate the performance of an IEEE 802.15.4 network in a noisy environment, which is typical of urban deployments or factory deployments.
• We have studied several problems on linear networks, as the dynamic addressing of nodes, or the design of a MAC layer which can correctly manage the traffic increase as packets become closer to their destination (this increase being caused by traffic aggregation). Linear networks occur in many real applications, such as pipeline monitoring, river monitoring, or when interconnecting distant hotspots.
• We work on the management of the mobility in wireless networks. We have proposed improvements to the RPL routing protocol, in order to react quickly to topology changes or to node mobility.
• We work on the distant take-over of mobile robots. Such robots are meta-sensors (that is, a mobile node equipped with several sensors) having throughput and latency constraints. We are studying a Wi-Fi solution based on ad hoc mode in order to establish reliable links, and to guarantee quality of service for the applications exploiting these robots.
• We study hybrid communications within C-ITS and 5G networks. We aim at designing techniques that make the optimal choice of the transmission technology within a heterogeneous network composed of several technologies such as LTE, Wi-Fi, ITS-G5, ZigBee, etc.
• We study the security of 5G networks. We take into account the limited capacity of nodes when choosing cryptographic operations, and when deciding how they will be executed.
• We are interested in authenticated key exchange in a 5G network composed of connected objects with limited computational capacities.
• We study the design of distributed algorithms for fleets of robots that are capable of moving, and are equipped with lights of several colors, but with otherwise very limited capabilities (no memory, no communication mean except the lights, only front vision, no compass). Our aim is to explore a discrete 2D or 3D space, while using as few robots and as few colors as possible.
• We design distributed protocols for highly dynamic networks. Our aim is to determine the smallest set of assumptions on the dynamicity while guaranteeing robustness properties. Examples of assumptions on dynamicity are: Should the network be always connected? Should links be available periodically? Examples of robustness properties are: Is the network robust against node crashes? Is the network robust against memory corruption? Is the network robust against unreliable communications?

We organize a monthly seminar: https://sancy.iut.uca.fr/~durand/seminaire/