Biometrics refers to the automated recognition of individuals based on their behavioural or biological characteristics. In spite of their numerous advantages over traditional authentication systems based on PINs or passwords (e.g., biometric characteristics cannot be lost or forgotten), biometric systems are vulnerable to external attacks and can leak privacy. Presentations attacks (PAs) - impostors who manipulate biometric systems by masquerading as other people - are serious threats to security. Privacy concerns involve the use of personal and sensitive biometric information, as classified by the GDPR, for purposes other than those intended. Vulnerabilities to PAs and privacy leakage are unacceptable and have hindered the deployment of biometric technology in commercial applications. The biometrics community has responded with presentation attack detection (PAD) technologies and privacy preservation mechanisms (biometric template protection schemes, BTP). Even though the latest PAD technologies are largely successful in protecting biometrics systems from known forms of PA, they tend to lack generalisation to different forms of attacks. The standard approach to privacy preservation involves some form of encryption or irreversible transformations, though the most recent fully homomorphic algorithms are general computationally prohibitive. Multi-biometric systems, explored extensively as a means of improving recognition reliability, also offer potential to improve PAD generalisation. Multi-biometric systems offer natural protection against spoofing since an impostor is less likely to succeed in fooling multiple systems simultaneously. For the same reason, previously unseen PAs are less likely to fool multi-biometric systems protected by PAD. Unfortunately, each sub-system in a multi-biometric approach to recognition has potential to leak privacy. Multi-biometric systems only compound the need for computationally prohibitive privacy preservation. RESPECT, a Franco-German collaborative project, will explore the potential of using multi-biometrics as a means to defend against diverse PAs and improve generalisation while still preserving privacy. Central to this idea is the use of (i) biometric characteristics that can be captured easily and reliably using ubiquitous smart devices and, (ii) biometric characteristics which facilitate computationally manageable privacy preserving, homomorphic encryption. The research will focus on characteristics readily captured with consumer-grade microphones and video cameras, specifically face, iris and voice. Further advances beyond the current state of the art involve the consideration of dynamic characteristics, namely utterance verification and lip dynamics. The core research objective will be to determine which combination of biometrics characteristics gives the best biometric authentication reliability and PAD generalisation while remaining compatible with computationally efficient privacy preserving BTP schemes.

Identifying, extracting, structuring, and storing knowledge are major knowledge management tasks. They constitute important challenges for organizations, partly because knowledge is scattered across different types of sources (e.g. databases, spreadsheets, textual documents) and heterogeneously represented. For instance, a large number of data repositories within companies as well as on Open Data portals are represented in the form of tabular data (spreadsheets) whereas PDF reports or Web pages frequently mix texts and tables. Hence, there is a need to structure and reconcile such scattered knowledge, which can be achieved by automatically extracting knowledge from heterogeneous sources to build and refine knowledge graphs. Such an extraction and refinement process enables a mutual correction and completion between texts, tables, and knowledge graphs. Interestingly, texts and tables may be related in the same document or across documents and complement one another, a complementarity that is little used so far. From these observations, the ECLADATTA project aims at leveraging this complementarity between tables, texts, and knowledge graphs to propose an end-to-end process that builds corpora of related texts and tables, and performs a joint knowledge extraction and reconciliation to enrich or update a knowledge graph. Such a process raises several issues that will be tackled by the ECLADATTA project. For example, assessing the relatedness between knowledge graphs, texts, and tables requires to delimit the exact text portion associated with a table and to compare atomic information taking into account temporal validity or aggregates such as means or sums. This process will be evaluated on collections of public documents collected from the web (e.g Wikimedia projects such as Wikipedia, with the ambition of scaling to large corpora such as the Common Crawl) to enrich publicly available knowledge graphs such as Wikidata.

TRUST focuses on personal data protection measures to meet the objectives of the RGPD but also the texts in preparation such as the "Data Act" or the "Data Governance Act". We propose to study and develop new security solutions, based on advanced cryptography, for use cases involving the reuse of personal data. These use cases will present various configurations in terms of actors, type of data and processing, opening the way to different technical and legal issues. We thus seek to anticipate legal evolutions and prepare technical architectures to allow the reuse of personal data in compliance with the various legal frameworks.

Thanks to the exponential growth of Internet, citizens have become more and more exposed to personal information leakages in their digital lives. This trend began with web tracking when surfing the Internet with our computers. Then the advent of smartphones, our personal assistants always connected and equipped with many sensors, further reinforced this tendency. And today the craze for “quantified self” wearable devices, for smart home appliances or for other connected devices enable the collection of potentially highly sensitive personal information in domains that were so far out of reach. However, little is known about the actual practices in terms of security, confidentiality, or data exchanges. The end-user is therefore prisoner of a highly asymmetric system. This has important consequences in terms of regulation, sovereignty, and leads to the hegemony of GAFA. Security, transparency and user control are three key properties that should be followed by all the stakeholders of the smartphone and connected devices ecosystem. Recent scandals show that the reality is sometimes at the opposite. The DAPCODS project gathers four renowned research teams, experts in security, privacy and digital economy. They are seconded by CNIL, the French data protection agency. The project aims at contributing along several axes: 1- by analyzing the inner working of a significant set of connected devices in terms of personal information leaks. This will be made possible by analyzing their data flows (and associated smartphone application if applicable) from outside (smartphone and/or Wifi network) or inside, through on-device static and dynamic analyses. New analysis methods and tools will be needed, some of them leveraging on previous works when applicable; 2- by studying the device manufacturers' privacy policies along several criteria (e.g. accessibility, precision, focus, privacy risks). In a second step, their claims will be compared to the actual device behavior, as observed during the test campaigns. This will enable an accurate and unique ranking of connected devices; 3- by understanding the underlying ecosystem, from the economical viewpoint. Data collected will make it possible to define the blurred boundaries of personal information market, a key aspect to set up an efficient regulation; 4- finally by proposing a public website that will rank those connected devices and will inform citizens. We will then test the impact of this information on the potential change of behavior of stakeholders. By giving transparent information of hidden behaviors, by highlighting good and bad practices, this project will contribute to reduce the information asymmetry of the system, to give back some control to the end-users, and hopefully to encourage certain stakeholders to change practices.

More than one billion homes worldwide still lack a broadband Internet connection. In addition, power consumption related to telecommunication network is constantly increasing following data traffic exponential growth. EEMW4FIX ambition is to offer reliable, high data rate and low-power access to end-users by using advanced antenna architectures for future wireless backhauls and Fixed Wireless Access (FWA). To this end, EEMW4FIX aims at developing innovative low-profile, high-gain, and steerable beam smart antenna, using 3D-printed flat lens. EEMW4FIX will address 3 main unresolved challenges needed for mmW FWA: - Drastically improving system energy efficiency of antenna system, RF front-end and beamforming algorithms. Back of the envelope calculations suggest that the EEMW4FIX approach can achieve a factor 10 of reduction in power consumption by combining 4 ingredients. The collimating gain provided by lens approach allows to reduce transmit power and increase reception sensitivity proportionally. The Massive MIMO system is realized via a lens antenna and beam space processing, which leads to beamforming algorithms with highly reduced computational complexity (which is normally cubic in the number of antennas). In addition, the number of activated antennas at any time in the feeding array is small compared to a classical antenna array in which all antenna elements are activated, leading to a significant reduction in the number of RF front-ends. Finally, the RF front-end thermal power will be harvested using integrated Peltier cells, further increasing the global system power efficiency. - Design of low-profile highly-directive steerable beam antenna. Most solutions available today exhibit a limited number of switched beam angles, using transmitarray or conventional bulk lenses without any fine beam tuning capability. In EEMW4FIX, a flat full dielectric multifocal lens will be optimized to spatially couple with a steerable phased array to obtain a high and quasi-constant directivity for all steered angles while ensuring extremely low spillover loss. This lens will be monolithically integrated inside a radome by additive manufacturing. Such concept has never been studied. - Extension for dual-band operation. Using multiple frequency bands enables operators to capitalize on the massive bandwidth available in mmW (37.75-40 GHz and 58-64 GHz) for upgrading the last kilometers access network. As a proof-of-concept, the 3D-printed lens of EEMW4FIX antenna will be designed for dual-band operation. Such capabilities are currently not available. EEMW4FIX gathers two academic partners (LEAT and Eurecom), one innovative SME (EV-Technologies) and two large companies (Orange and Thales), all selected for their complementary expertise.