Activities like mining exploration, port or coastal surveillance are increasingly carried out by swarms of drones controlled semi-automatically. The complexity of their networks makes them vulnerable to cyberattacks, however.

To secure the data contained and transmitted between drones, the ENSTA Bretagne/Lab-STICC Software/Hardware And unKnown EnviRonment Interactions (SHAKER) team launched the DISPEED* project with the AID in September 2022. Its goal? “To develop an intrusion detection system (IDS) factoring in the resources which each of the drones in the swarm needs in terms of energy and calculating capacity,” explains Camélia Slimani, a post-doctoral student at ENSTA Bretagne and a member of this team. 

The most widespread IDSs leverage machine learning algorithms which require significant memory and computing power. Not all types of drones have the same processing capacities though (processors, memory, storage), which affects their cybersecurity performance. “The challenge is to come up with an execution model which strikes a relevant trade-off between swift detection and energy use depending on the criticality of the attack and state of the system and the mission,” the researcher clarified.

The research team initially conducted an energy use and performance study of several existing IDSs before drawing up an appropriate execution strategy for the missions chosen for a population of drones operating autonomously. 

* Project “Détection d’Intrusion et compromis Sécurité / Performance / Energie, Etude pour les meutes de Drones” (“Intrusion Detection and Security / Performance / Energy tradeoff, a Study for Drone swarms”) financed by the Ministry for the Armed Forces Defense Innovation Agency (AID).

• Objective of this project: to develop methods and tools for modeling data access profiles in an efficient and non-intrusive manner then use the models established to develop strategies for optimizing the energy consumption of compute nodes during the data access stage.
• Financed by: Atos
• Led by: Jalil Boukhobza, research professor at ENSTA Bretagne/Lab-STICC (SHARP department, SHAKER team)

In performance terms, storage systems represent one of the most significant weak links in an IT system, particularly for applications which process large amounts of data such as in the field of high-performance computing (HPC).  

The emergence of new storage technologies is an opportunity to reduce the performance gap between working memory and storage, as well as energy consumption. 
These technologies are deployed at various levels:

- storage memory (e.g. 3DxPoint),
- its interface (e.g. NVMe),
- or its software management (e.g. object store).

These technologies imply significant growth in the complexity of storage management in order to meet the service quality requirements of applications. 
 
 

Objective: outline a model and strategies for running intrusion detection systems embedded on drones equipped with heterogeneous architectures and striking a relevant compromise according to the attack criticality and state of the system and mission, between detection speed/energy consumption.

• Financed by: AID
• Partners: FORTH Greece, Naval Group, UBO
• Led by: Jalil Boukhobza, research professor at ENSTA Bretagne/Lab-STICC (SHARP department, SHAKER team)

The distributed operation of fleets of drones during missions makes them vulnerable to diverse attacks that it is crucial to detect. Embedded in these drones are hardware components (computing and storage) with heterogenous computing power and energy consumption for performing the various tasks necessary for their mission.

The project sets out to develop models, strategies and tools for optimizing the energy cost of intrusion detection on a fleet of drones or any other cooperative system with major energy or hardware capacity constraints. These systems operate in cooperation to accomplish a joint mission. The network load therefore varies enormously depending on the context of the mission, which means that the intrusion detection system does not need to be run continuously on equipment requiring significant hardware capacity or consuming a considerable share of the system’s energy.

The aim of the project is thus to study and analyze how the performance of the IDS can be adapted using various hardware components depending on this network load and the context of the mission.

4 challenges underpin the project:

• Challenge 1: modeling the execution environment
• Challenge 2: setting up an assessment platform
• Challenge 3: designing a configuration selection strategy based on multi-objective optimization tooling
• Challenge 4: implementing an inter-drone balancing/offset computing system for reducing energy consumption.

Objective: design effective data positioning systems on multi-tiered storage architectures in the field of high-performance computing.

• Financed by: CEA
• Led by: Jalil Boukhobza, research professor at ENSTA Bretagne/Lab-STICC (SHARP department, SHAKER team)

In performance terms, storage systems represent one of the most significant weak links in an IT system, particularly for applications which process large amounts of data. The emergence of new storage technologies is an opportunity to reduce the performance gap between working memory and storage. These technologies, deployed at the level of storage memory (e.g. 3DxPoint), its interface (e.g. NVMe) and even its software management (e.g. object store), imply significant growth in the complexity of storage management. In addition, amid the "big Data" boom, more and more applications are processing huge amounts of data, and present different levels of criticality.

Against this backdrop, we intend to study and come up with new data positioning strategies with different levels of criticality on heterogenous, geo-distributed storage systems. As part of this project, we will explore several techniques including machine learning, reinforcement learning and optimization methods, to guarantee effective online data positioning.

The world is growing increasingly connected today, and industry is no exception. 

Factories are becoming equipped with ever more connected objects (sensors, actuators, etc.) which provide real-time monitoring of the proper operation of production machinery. The data that these objects collect is then sent to a hardware device known as a gateway, which can analyze this data and detect any anomalies. The whole of this system makes it possible to carry out predictive maintenance. Communication between this gateway and the objects is vulnerable, however. For the data is transferred between remote-controlled objects and the gateway via wireless networks (Wi-Fi, Bluetooth), which can be subject to cyberattacks.

Innovative technologies

The national TrustGW project was launched in 2021 to ensure the secure operation of a factory. It brings ENSTA Bretagne together alongside Université Bretagne Sud, Irisa Rennes and IETR Rennes.

Its goal? Implement a cybersecurity solution to protect data when it is being collected by the connected object, transferred to the gateway and analyzed.

“The project is original in that it entails development of a reconfigurable gateway capable of swiftly processing very large datasets from several sensors using different wireless networks (Wi-Fi, Bluetooth, etc.),” explains Pascal Cotret, lecturer at Lab-STICC, in charge of developing a prototype for industry. With that in mind, the researchers are particularly using FPGA electronic components for the rapid execution of cryptographic and hashing algorithms for securing data on RISC-V open-source processors.

They are also less vulnerable to cyberattacks as they can be reprogrammed over time. “To make gateway datasets more secure, we partition them depending on their characteristics, separating out the data depending on whether it comes from Wi-Fi- or Bluetooth-connected objects.”

A number of industrial players are already showing an interest in the technology developed through the TrustGW project, for the purposes of upgrading existing industrial facilities or installing new protected networks within their future factories.

• Research program name: "Protection against intellectual property theft in electronics: code obfuscation techniques for HLS synthesis chains in SaaS mode"
• began at the end of 2017
• financed by the Brittany region
• led by Jean-Christophe Le Lann (research professor)
• partner: Politechnico di Milano (public scientific-technological university which trains engineers, architects and industrial designers, based in Milan)
• thesis by Hannah Boenning Badier, defended in 2021

The PhD student and team explored various promising avenues in terms of obfuscation (security through obscurity), which involves incorporating a discreet "tattoo" into a computer code using artificial intelligence techniques.

Trojan horses were also designed as part of this project in a bid to more clearly understand the implications of such hardware threats and to consider countermeasures at the time of design.

Jean-Christophe Le Lann: "This original project is set to continue with support from the Cyber Center of Excellence. It is of interest to the French Defense, Procurement and Technology Agency (DGA) and will be the subject of a second thesis beginning in 2022."
 

Against a fiercely competitive global economic backdrop, the aeronautical industry is one of France’s strengths. With a fabric of small, medium and large companies, the French aeronautics sector is the only one, along with the United States, that is fully capable of developing, producing and marketing civil and military aircraft. The French Government has developed a plan for supporting the aeronautics sector, designed to protect French expertise and know-how, while delivering the far-reaching changes needed to achieve the energy transition. The strategy is focused on the green transition and lowering carbon emissions in air transport.

The French aeronautical industry’s expertise on its products, programs and interactions within its value chain is widely recognized. For all that, it must contend with a growing number of challenges if it is to become more proficient in its design and development cycles and more efficient in its engineering activities and ensure that the performances of its products and support systems continually improve. It also needs to take technological innovations more swiftly on board and take advantage of the opportunities offered by new information technologies. Given these challenges, there is an inevitable need for radical changes to engineering methods within the French aeronautical industry, and this is where the ONEWAY project comes in. 

The project began in May 2021, for an 18-month period, with a budget of €48m. It brought together 14 partners: Airbus, Dassault Aviation, Liebherr Aerospace, Safran Electrical & Power, Safran Aerotechnics, Thales, Altran Technologies, Cap Gemini, Sopra Steria, CIMPA, PragmaDEV, IMT Mines Ales, Université de Rennes 1 and ENSTA Bretagne.

ENSTA Bretagne helped to define a digital capacity for supporting decisions regarding launch, then control and management of a Product Development Plan (PDP). The PDP seeks to predict and control the best date for a product and its industrial system to be brought to the market, as well as the expected production ramp-up stage. This has become crucial for the competitiveness of the French aeronautical industry.

Thanks to the experience of ENSTA Bretagne’s Processes for Safe and Secure Software and Systems (P4S) team on federating complex software systems, the development of formal semantics and analytical algorithms, an equipped PDP modeling framework has been established. The tool developed allows for a detailed capture of the business specifics, industrial-scale simulation of the development process and validation of the models built through formal verification methods. 

For ENSTA Bretagne, the two main implications in the ONEWAY project concern the formal verification and validation of the PDP. Project outcomes:

1. Extension of the OBP2 model-checker with statistical exploration algorithms for massive testing on industry-derived models;
2. Improvement of the layer of expression of formal properties associated with the system requirements or Top Program Objectives;
3. Invention of a modular strategy for the formal verification of time-bound systems, based on the PDP’s formal semantics without the need for costly model processing procedures. 

• Cyber-security modelling and analysis framework" research project: Developing a cohesive framework for the specification, formalization and analysis of secure software and hardware architecture
• in progress since December 2020
  funded by the AID (Defense Innovation Agency)
• led by Raul Mazo Pena, a research professor at ENSTA Bretagne / Lab-STICC (SHARP department, P4S team)

To find out more: read the article on this program

It is still early days for the "Security by Design" approach and significant R&D efforts will be required for its use to become systematic and widespread. That’s the aim of this groundbreaking project, which is in some ways opening up a whole new engineering discipline by outlining a new vision. To take up this challenge, the project sets out to create a cohesive, overarching theory, with systematic design tools, techniques and methods.

• Project funded by the Brittany Region and FEDER
• started at the end of 2019 until 2022
• 3 partners: KEREVAL, Mobility Tech Green and ENSTA Bretagne
• led by: Joël Champeau, a research professor at ENSTA Bretagne, UMR (joint research unit) Lab-STICC (SHARP department, P4S team

The project sets out to develop products and services embedded in connected vehicles, as well as associated off-board services.

These on-board services will have gone through a secure development process.

ENSTA Bretagne’s contribution involves developing a design methodology and tooling of cybersecurity tests specially geared towards "connected vehicles".

This method will need to range from the system level, factoring in the security requirements, to the communication modules of the embedded calculator.

The expected results of the project are the development of such new mobility services as fleet management, development of a CyberLab to run the security tests of the services and a methodological support grounded in a formal verification of the security requirements.

...

This study is aimed at modeling, simulating and optimizing aerial drones’ physical communication channels (from the transmitter to the receiver), with account taken of operational contexts. 

It is the second stage in a vast research program commissioned by the DGA, the aim being to optimize the quality of communications between an aerial drone and its operator based on land, or between several drones and their operator. Tools for calculating missions and controlling the drones will have to take the operational contexts into account for that, not least such natural features as terrain or dense vegetation, which limit wave propagation.

This path loss coefficient was modeled during the first study. The new study has several additional objectives. 

First of all, finding the best routes for maintaining the highest possible path coefficient. The researchers are developing effective optimization techniques in that respect, which incorporate radar wave propagation models (mathematical theories of optimization).

Other parameters taken into account include flight management and drone orientation. These also affect radio frequency communications and are factored into models using AI algorithms (reinforcement learning).

Finally, mission calculation methods may bring two or more drones into play to ensure this optimization of communications. In that case, the researchers study the extent to which the algorithms developed are capable of coping with the combinatorial explosion inherent in the multi-drone context.

• thesis subject: “Towards Effective Strategies for Data Collection and Decision-Making for Sensor Networks with Limited Resources”
• defended in December 2021
• keywords: Sensor Networks, Energy Efficiency, Decision-making, Data Reduction, Spatio-Temporal Correlation, Planning Strategies.

The high potential of sensors is hampered by two major technological hurdles: the limited resources of their batteries and the difficulty collecting large datasets in real-time. This theme aims at outlining various mechanisms for collecting and analyzing data to overcome these challenges, on the basis of sensor network architecture (clustering).

The results obtained during testing have demonstrated the effectiveness of these mechanisms in terms of energy consumption, data precision and scope, and improve the performances of sensor networks.

The mechanisms proposed operate at three network levels. They aim at reducing the quantity of data disseminated through the network while protecting the integrity of information.

• At sensor level, we outline methods for the prediction, aggregation and compression of data based respectively on the algorithms of Newton Forward Difference, divide-and-conquer and similar data elimination, with a view to reducing the raw data collected by each sensor.
• At “Cluster Head” level (i.e. leading node in the sensor network), we outline new techniques for clustering, fusion, intermediate aggregation and scheduling aimed at looking for correlations between the neighboring nodes and eliminating existing redundant data before sending the data to the sink.
• At sink level (node/gateway of the sensor network), we introduce effective decision-making models based on a score table, allowing end users to analyze the data and make a fitting decision.

These programs are conducted with the military hospital and teaching hospital (CHRU) in Brest.

• Acquisition and processing of electrocardiograms of a fetus and its mother using wireless sensors.
• Characterization and classification of deep vein thrombosis (blood clot).
• Use of EEG (electroencephalogram) and EMG (electromyogram) signals to control a wheelchair for a paraplegic.
• Use of EOG (electro-oculogram) signals to activate and browse a web page by a paralyzed person: produce a wireless ECG sensor and simulator for the Faculty of Medicine.

By estimating the characteristics of the transmission channel to improve information transmission and protection.

• Game theory to develop protocols for a tactical cognitive radio.
• Smart Antenna & Beamforming: the antenna has to adapt to its environment automatically.
• Internet of Things and wireless network problems associated with the coexistence of machine-to-machine (M2M) and human-to-human (H2H) communications.

These algorithms use no-threshold measures, in scenarios where conventional approaches fail, owing to low signal-to-noise ratios (SNR) or restrictive environments.

An original algorithm has been developed in partnership with Diades Marine, as part of the ADEME e-PANEMA project (e-Positioning and Navigational Aid in the Marine Environment) and won the IEEE Antennas and Propagation Society award in 2019.

It was developed by using particle filters for the detecting and tracking of targets; its performances were assessed using actual radar data.

DOREDO is a system for detecting and locating obstacles and objects, which can be embedded on medium-endurance drones, warning of any potential collision route with other aircraft (of the light aircraft, helicopter or recreational drone type for example). 

The system allows for secure flight, with comparable characteristics to current airliners thanks to the removal of such technological hurdles as miniaturization and robustness regarding air or land clutter.

>> DOREDO is a first step towards the flight of medium-endurance drones within non-segregated airspace.

Funded by: DGA. Partners: CESTIM, CNAM Paris

Scalable deep-learning techniques for the detection and recognition of targets from heterogeneous data.

Funded by: DGA, AID, I2R

Characterizing the atmosphere and sea surface interactions for the deployment of offshore wind in the Gulf of Lion.

• OBJECTIVE: To support the development of offshore wind energy in the French Mediterranean coastal areas by providing a database of high-resolution observations of wind, wave and current fields as well as a new numerical tool for the modelling of metocean conditions in the Gulf of Lion.
• Timetable: 2020-2023
• Funded by: France Energies Marines
• 15 national and international partners

>> Find out more

...

...