Publications

2023

• Microscopic characteristics of copper wires with short-circuit molten marks in electrical fire
  
  • Xu Na
  • Ding Ning
  • Liu Long
  • Zaïri Fahmi
  • Tian Linan
  • Guo Weimin
  • Li Fuwei
  • Li Enxia
  • Chen Lizong
  • Li Nan
  Materials Testing, 2023, 65 (6), pp.844-854. Abstract Electrical fire accidents occurred frequently all over the world in recent years, which had caused a large number of casualties and property losses. The main cause of electrical fire is electrical wire short circuit fault. Investigations of the characteristic and influence factors of the molten marks play an important role to determine the cause of fire. In the present work, microscopic characteristics of copper wires with short-circuit molten marks in electrical fire were investigated. Primary and secondary short-circuit molten marks (PMMs and SMMs) of copper wires were prepared by simulation test. The micro-morphology, microstructure and chemical composition of molten marks were observed and compared. Based on the quantitative characterization, effects of different overload currents and heating temperatures on the microstructure of molten marks were analyzed. The results would provide fundamental information for cause identification of the electrical fire. (10.1515/mt-2022-0423)
  DOI : 10.1515/mt-2022-0423
• Fingerprinting and Building Large Reproducible Datasets
  
  • Lefeuvre Romain
  • Galasso Jessie
  • Combemale Benoit
  • Sahraoui Houari
  • Zacchiroli Stefano
  , 2023, pp.27-36. Obtaining a relevant dataset is central to conducting empirical studies in software engineering. However, in the context of mining software repositories, the lack of appropriate tooling for large scale mining tasks hinders the creation of new datasets. Moreover, limitations related to data sources that change over time (e.g., code bases) and the lack of documentation of extraction processes make it difficult to reproduce datasets over time. This threatens the quality and reproducibility of empirical studies. In this paper, we propose a tool-supported approach facilitating the creation of large tailored datasets while ensuring their reproducibility. We leveraged all the sources feeding the Software Heritage append-only archive which are accessible through a unified programming interface to outline a reproducible and generic extraction process. We propose a way to define a unique fingerprint to characterize a dataset which, when provided to the extraction process, ensures that the same dataset will be extracted. We demonstrate the feasibility of our approach by implementing a prototype. We show how it can help reduce the limitations researchers face when creating or reproducing datasets. (10.1145/3589806.3600043)
  DOI : 10.1145/3589806.3600043
• Proving the Convergence to Limit Cycles using Periodically Decreasing Jacobian Matrix Measures
  
  • Jerray Jawher
  • Fribourg Laurent
  , 2023. Methods based on "(Jacobian) matrix measure" to show the convergence of a dynamical system to a limit cycle (LC), generally assume that the measure is negative everywhere on the LC. We relax this assumption by assuming that the matrix measure is negative "on average" over one period of LC. Using an approximate Euler trajectory, we thus present a method that guarantees the LC existence, and allows us to construct a basin of attraction. This is illustrated on the example of the Van der Pol system.
• Unbalanced CO-Optimal Transport
  
  • Tran Quang Huy
  • Janati Hicham
  • Courty Nicolas
  • Flamary Rémi
  • Redko Ievgen
  • Demetci Pinar
  • Singh Ritambhara
  Proceedings of the AAAI Conference on Artificial Intelligence, Association for the Advancement of Artificial Intelligence, 2023, 37 (8), pp.10006-10016. Optimal transport (OT) compares probability distributions by computing a meaningful alignment between their samples. COoptimal transport (COOT) takes this comparison further by inferring an alignment between features as well. While this approach leads to better alignments and generalizes both OT and Gromov-Wasserstein distances, we provide a theoretical result showing that it is sensitive to outliers that are omnipresent in real-world data. This prompts us to propose unbalanced COOT for which we provably show its robustness to noise in the compared datasets. To the best of our knowledge, this is the first such result for OT methods in incomparable spaces. With this result in hand, we provide empirical evidence of this robustness for the challenging tasks of heterogeneous domain adaptation with and without varying proportions of classes and simultaneous alignment of samples and features across single-cell measurements. (10.1609/aaai.v37i8.26193)
  DOI : 10.1609/aaai.v37i8.26193
• Engineering Design of Integrated III-V/SiN Quantum-well and Quantum-dot Lasers
  
  • Alkhazraji Emad
  • Chow Weng W
  • Grillot Frederic
  • Bowers John E
  • Madaras Scott E
  • Gehl Michael
  • Skogen Erik
  • Wan Yating
  , 2023. Spectrally-pure lasers are paramount in various fields. Progression to Hz-level lasing linewidth in III-V/SiN lasers with quantum-dot active regions is predicted here. Using parametric studies, one can produce timely results in engineering designs.
• Private free-space transmission based on chaos synchronisation in the 8-14 µm atmospheric transparency window
  
  • Didier Pierre
  • Zaminga Sara
  • Spitz Olivier
  • Awwad Elie
  • Maison Gregory
  • Carras Mathieu
  • Grillot Frédéric
  , 2023.
• Flexible Analog-to-Feature converter for wireless smart healthcare sensors
  
  • Manokhin Mikhail
  • Chollet Paul
  • Desgreys Patricia
  , 2023. Analog-to-Feature (A2F) conversion based on Non-Uniform Wavelet Sampling (NUWS) has demonstrated the ability to drastically reduce the energy consumption in wireless sensors while employed for electrocardiogram (ECG) anomaly detection. The underlying idea is to extract relevant features from the analog signal and perform the classification in the digital domain. We adopt the same approach for a human activity recognition (HAR) task, considered as a second application for a proposed generic A2F converter. By extracting only 16 features from the inertial signals of the UCI-HAR data set and using these features as inputs for a simple Neural Network, we achieved an 87.7% accuracy in multiclass classification. From the simulation results, we defined the relevant features and the hardware specifications required for a complete circuit design and chip fabrication.
• Time-frequency quantum metrology
  
  • Fabre N.
  , 2023, pp.paper eb_7_6. Hong-Ou-Mandel interferometry takes advantage of the quantum nature of two-photon interference to increase the resolution of precision measurements of time delays. Relying on few-photon probe states, this approach is applicable also in cases of extremely sensible samples and it achieves attosecond-scale resolution, which is relevant to cell biology and two-dimensional materials. Here, we theoretically analyze how the precision of Hong-Ou-Mandel interferometers can be significantly improved by engineering the spectral distribution of two-photon probe states. In particular, we assess the metrological power of different classes of biphoton states with non-Gaussian time-frequency spectral distributions, considering the estimation of both time and frequency shifts. We find that grid states, characterized by a periodic structure of peaks in the chronocyclic Wigner function, can outperform standard biphoton states in sensing applications.
• Data Compression with Private Local Decodability
  
  • Tchamkerten Aslan
  • Vatedka Shashank
  • Chandar Venkat
  , 2023. Classical compression schemes suggest that message symbols cannot be privately decoded; if a string X n is encoded into a codeword C nR at a non-trivial rate R, then the decoding of an individual symbol X i reveals information about the rest of the symbols X n \X i .While this holds for virtually all lossless compression schemes, it is shown that this need not be the case. This paper proposes a lossless compression scheme for bit strings with the following properties. For any sufficiently small p > 0, it encodes each length-n bit string of Hamming weight at most np into a binary codeword of length O(nplog21p) such that the subset of compressed bits that need to be probed in order to decode a particular message bit reveals no additional information about the other message bits.
• Integrated Communication and Receiver Sensing with Security Constraints on Message and State
  
  • Ahmadipour Mehrasa
  • Wigger Michèle
  • Shamai Shlomo
  , 2023, pp.2738-2743. We study the state-dependent wiretap channel with non-causal channel state informations at the encoder in an integrated sensing and communications (ISAC) scenario. In this scenario, the transmitter communicates a message and a state sequence to a legitimate receiver while keeping the message and state-information secret from an external eavesdropper. This paper presents a new achievability result for this doubly-secret scenario, which recovers as special cases the best-known achievability results for the setups without security constraints or with only a security constraint on the message. The impact of the secrecy constraint (no secrecy-constraint, secrecy constraint only on the message, or on the message and the state) is analyzed at hand of a Gaussian-state and Gaussian-channel example (10.1109/ISIT54713.2023.10206763)
  DOI : 10.1109/ISIT54713.2023.10206763
• The interplay between error, total variation, alpha-entropy and guessing: Fano and Pinsker direct and reverse inequalities
  
  • Rioul Olivier
  Entropy, MDPI, 2023, 25 (7), pp.978. Using majorization theory via “Robin Hood” elementary operations, optimal lower and upper bounds are derived on Rényi and guessing entropies with respect to either error probability (yielding reverse-Fano and Fano inequalities) or total variation distance to the uniform (yielding reverse-Pinsker and Pinsker inequalities). This gives a general picture of how the notion of random- ness can be measured in many areas of computer science.
• Mixing a covert and a non-covert user
  
  • Bounhar Abdelaziz
  • Sarkiss Mireille
  • Wigger Michèle
  , 2023, pp.2577-2582. This paper establishes the fundamental limits of a two-user single-receiver system where communication from User 1 (but not from User 2) needs to be undetectable to an external warden. Our fundamental limits show a tradeoff between the highest rates (or square-root rates) that are simultaneously achievable for the two users. Moreover, coded time-sharing for both users is fundamentally required on most channels, which distinguishes this setup from the more classical setups with either only covert users or only non-covert users. Interestingly, the presence of a non-covert user can be beneficial for improving the covert capacity of the other user. (10.1109/ISIT54713.2023.10206954)
  DOI : 10.1109/ISIT54713.2023.10206954
• Strong Converses for Memoryless Bi-Static ISAC
  
  • Ahmadipour Mehrasa
  • Wigger Michèle
  • Shamai Shlomo
  , 2023, pp.1818-1823. The paper characterizes the fundamental limits of integrated sensing and communication (ISAC) systems with a bi-static radar, where the radar receiver is located close to the transmitter and estimates or detects the state based on the transmitter’s channel inputs and the backscattered signals. Two models are considered. In the first model, the memoryless state sequence is distributed according to a fixed distribution and the goal of the radar receiver is to reconstruct this state-sequence with smallest possible distortion. In the second model, the memoryless state is distributed either according to P S or to Q S and the radar’s goal is to detect this underlying distribution so that the missed-detection error probability has maximum exponential decay-rate (maximum Stein exponent). Similarly to previous results, our fundamental limits show that the tradeoff between sensing and communication solely stems from the empirical statistics of the transmitted codewords which influences both performances. The main technical contribution are two strong converse proofs that hold for all probabilities of communication error ϵ and excess-distortion probability or false-alarm probability δ summing to less than 1, ϵ+δ < 1. These proofs are based on two parallel change-of-measure arguments on the sets of typical sequences, one change-of-measure to obtain the desired bound on the communication rate, and the second to bound the sensing performance (10.1109/ISIT54713.2023.10207009)
  DOI : 10.1109/ISIT54713.2023.10207009
• Maximal leakage of masked implementations using Mrs. Gerber's lemma for min-entropy
  
  • Béguinot Julien
  • Liu Yi
  • Rioul Olivier
  • Cheng Wei
  • Guilley Sylvain
  , 2023. A common countermeasure against side-channel attacks on secret key cryptographic implementations is dth-order masking, which splits each sensitive variable into d + 1 random shares. In this paper, maximal leakage bounds on the probability of success of any side-channel attack are derived for any masking order. Maximal leakage (Sibson’s information of order infinity) is evaluated between the sensitive variable and the noisy leakage, and is related to the conditional “min-entropy” (Arimoto’s entropy of order infinity) of the sensitive variable given the leakage. The latter conditional entropy is then lower- bounded in terms of the conditional entropies for each share using majorization inequalities. This yields a generalization of Mrs. Gerber’s lemma for min-entropy in finite Abelian groups.
• Feedback Increases the Capacity of Queues with Finite Support Service Times
  
  • Tchamkerten Aslan
  • Sahasranand K. R.
  , 2023, pp.7-12. In their "Bits Through Queues" paper, Anantharam and Verdú showed that if the service time is memoryless feedback does not increase capacity under a FIFO policy, and further conjectured that feedback increases capacity for all other service times. Towards this conjecture, a recent paper by Aptel and Tchamkerten provided a sufficient condition on the service time under which feedback increases capacity. While this condition yields examples of service times for which feedback is helpful, it does not provide explicit structural properties of such service times.In this paper, we consider the discrete-time setting and show that feedback increases capacity for any service time with finite support. We also show that the above sufficient condition is inconclusive for service times with infinite support. (10.1109/ISIT54713.2023.10206779)
  DOI : 10.1109/ISIT54713.2023.10206779
• Confidential Truth Finding with Multi-Party Computation (Extended Version)
  
  • Saadeh Angelo
  • Senellart Pierre
  • Bressan Stéphane
  , 2023. Federated knowledge discovery and data mining are challenged to assess the trustworthiness of data originating from autonomous sources while protecting confidentiality and privacy. Truth-finding algorithms help corroborate data from disagreeing sources. For each query it receives, a truth-finding algorithm predicts a truth value of the answer, possibly updating the trustworthiness factor of each source. Few works, however, address the issues of confidentiality and privacy. We devise and present a secure secret-sharing-based multi-party computation protocol for pseudo-equality tests that are used in truth-finding algorithms to compute additions depending on a condition. The protocol guarantees confidentiality of the data and privacy of the sources. We also present variants of truth-finding algorithms that would make the computation faster when executed using secure multi-party computation. We empirically evaluate the performance of the proposed protocol on two state-of-the-art truth-finding algorithms, Cosine, and 3-Estimates, and compare them with that of the baseline plain algorithms. The results confirm that the secret-sharing-based secure multi-party algorithms are as accurate as the corresponding baselines but for proposed numerical approximations that significantly reduce the efficiency loss incurred. (10.48550/arXiv.2305.14727)
  DOI : 10.48550/arXiv.2305.14727
• Piezoelectric system
  
  • Pillonnet Gaël
  • Morel Adrien
  • Manokhin Mikhail
  , 2023.
• Uplink tip-tilt-focus estimation at point ahead angle from off-axis LGS and on-axis downlink measurements for satcom applications
  
  • Lognoné Perrine
  • Conan Jean-Marc
  • Paillier Laurie
  • Rekaya Ben Othman Ghaya
  • Bonnefois Aurélie Montmerle
  • Vedrenne Nicolas
  , 2023. Reaching very high data rates in GEO Feeder optical uplinks is greatly impaired by the fading nature of the atmospheric channel. One of the current solutions to reduce the atmospheric turbulence impact on the signal, and thus improve the statistics of the received power aboard the satellite, is to apply a pre-compensation by adaptive optics (AO). In such configuration, the phase is measured on the downlink beam and applied to the uplink. However, due to the point-ahead anisoplanatism inherent to the geometry of the GEO Feeder, uplink correction efficiency is limited. As a result, the received power aboard the satellite still undergoes long and deep fades and the information signal is impaired. To improve the pre-compensation, one solution could be to use a laser guide star (LGS) to measure the phase at point ahead angle (PAA). However, this method does not currently allow to retrieve the Tip, Tilt and Focus modes which are however crucial modes to be corrected in order to improve the coupling statistics. We propose here a method extended from one we recently published [Lognone et al, OpticsExpress 2023], to estimate the Tip, Tilt and Focus at PAA based on an MMSE estimation relying on LGS high order measurements concatenated with phase and log-amplitude measurements performed on the on-axis downlink beam. We evaluate the phase estimation performance and the gain on the uplink coupled flux statistics. The new estimator shows to reduce the Tip, Tilt and Focus error variance by up to 70% of their initial value.
• Taming Fabry–Pérot resonances in a dual-metasurface multiband antenna with beam steering in one of the bands
  
  • Gonçalves Licursi de Mello Rafael
  • Lepage Anne Claire
  • Begaud Xavier
  Scientific Reports, Nature Publishing Group, 2023, 13 (1), pp.9871:1-9871:10. Abstract Metasurfaces are artificial materials that can provide properties not readily available in nature for the interaction with acoustic, elastodynamic, or electromagnetic waves. In Electromagnetics, metasurfaces allow particular functionalities to antennas, which in turn lately have been increasingly pushed to a multiband operation. To fully exploit metasurfaces’ capabilities, the use of a metasurface reflector and a metasurface superstrate surrounding a radiating element in multiband antennas is interesting. However, such topology generally creates multiple reflections inside the formed cavity, known as Fabry–Pérot resonances. Here we show that one should tame this phenomenon to use two parallel metasurfaces surrounding a planar radiating element. We present the conditions to obtain directive, multiband antennas under such circumstances. The concepts are validated with a compact device for 5G/4G/Wi-Fi 2.4/5/6E performing a beam steering in the 5G without disturbing the radiation patterns of the other bands. This device demonstrates that the functionalities of two metasurfaces may be exploited in a single design if the presented conditions are respected. We also anticipate our work to be a starting point for other studies in the wave domain. For example, compact, multiband, beam-steerable microphones or sonar transducers with two parallel metasurfaces could be investigated in the future. (10.1038/s41598-023-36828-4)
  DOI : 10.1038/s41598-023-36828-4
• Optimal Security Notion for Decentralized Multi-Client Functional Encryption
  
  • Nguyen Ky
  • Phan Duong Hieu
  • Pointcheval David
  , 2023, 13906, pp.336-365. Research on (Decentralized) Multi-Client Functional Encryption (or (D)MCFE) is very active, with interesting constructions, especially for the class of inner products. However, the security notions have been evolving over the time. While the target of the adversary in distinguishing ciphertexts is clear, legitimate scenarios that do not consist of trivial attacks on the functionality are less obvious. In this paper, we wonder whether only trivial attacks are excluded from previous security games. And, unfortunately, this was not the case. We then propose a stronger security notion, with a large definition of admissible attacks, and prove it is optimal: any extension of the set of admissible attacks is actually a trivial attack on the functionality, and not against the specific scheme. In addition, we show that all the previous constructions are insecure w.r.t. this new security notion. Eventually, we propose new DMCFE schemes for the class of inner products that provide the new features and achieve this stronger security notion. (10.1007/978-3-031-33491-7_13)
  DOI : 10.1007/978-3-031-33491-7_13
• Minimal Generating Sets for Semiflows
  
  • Memmi Gerard
  , 2023, LNCS-13910, pp.189-205. We discuss important characteristics of finite generating sets for $$\mathcal {F^{+}}$$F+, the set of all semiflows with non-negative coordinates of a Petri Net. We endeavor to regroup a number of algebraic results dispersed throughout the Petri Nets literature and also to better position the results while considering semirings such as $$\mathbb {N}$$N or $$\mathbb {Q^+}$$Q+ then fields such as $$\mathbb {Q}$$Q. As accurately as possible, we provide a range of new algebraic results on minimal semiflows, minimal supports, and finite minimal generating sets for a given family of semiflows. Minimality of semiflows and of support are critical to develop effective analysis of invariants and behavioral properties of Petri Nets. Main results are concisely presented in a table and our contribution is highlighted. We conclude with the analysis of an example drawn from the telecommunication industry underlining the efficiency brought by using minimal semiflows of minimal supports. (10.1007/978-3-031-35355-0_12)
  DOI : 10.1007/978-3-031-35355-0_12
• One-shot Unsupervised Domain Adaptation with Personalized Diffusion Models
  
  • Benigmim Yasser
  • Roy Subhankar
  • Essid Slim
  • Kalogeiton Vicky
  • Lathuilière Stéphane
  , 2023, pp.698-708. Adapting a segmentation model from a labeled source domain to a target domain, where a single unlabeled datum is available, is one the most challenging problems in domain adaptation and is otherwise known as one-shot unsupervised domain adaptation (OSUDA). Most of the prior works have addressed the problem by relying on style transfer techniques, where the source images are stylized to have the appearance of the target domain. Departing from the common notion of transferring only the target ``texture'' information, we leverage text-to-image diffusion models (e.g., Stable Diffusion) to generate a synthetic target dataset with photo-realistic images that not only faithfully depict the style of the target domain, but are also characterized by novel scenes in diverse contexts. The text interface in our method Data AugmenTation with diffUsion Models (DATUM) endows us with the possibility of guiding the generation of images towards desired semantic concepts while respecting the original spatial context of a single training image, which is not possible in existing OSUDA methods. Extensive experiments on standard benchmarks show that our DATUM surpasses the state-of-the-art OSUDA methods by up to +7.1%. The implementation is available at https://github.com/yasserben/DATUM (10.1109/CVPRW59228.2023.00077)
  DOI : 10.1109/CVPRW59228.2023.00077
• A Genetic Algorithm for a Spectre Attack Agnostic to Branch Predictors
  
  • Bourgeoisat Dorian
  • Sauvage Laurent
  , 2023. This paper explores the possibility of using a genetic algorithm to launch a Spectre attack on different branch predictors with the same binary. We focus on RISC-V architectures, specifically Sonic-BOOM, as they are the most likely to have a wide variety of branch predictors while sharing the same ISA. We show that our genetic algorithm is able to find a training and attack sequence to mount a Spectre Bounds Check Bypass attack on multiple branch predictors.
• An Information-Theoretic Approach to Collaborative Integrated Sensing and Communication for Two-Transmitter Systems
  
  • Ahmadipour Mehrasa
  • Wigger Michèle
  IEEE Journal on Selected Areas in Information Theory, IEEE, 2023, 4, pp.112-127. This paper considers information-theoretic models for integrated sensing and communication (ISAC) over multi-access channels (MAC) and device-to-device (D2D) communication. The models are general and include as special cases scenarios with and without perfect or imperfect state-information at the MAC receiver as well as causal state-information at the D2D terminals. For both setups, we propose collaborative sensing ISAC schemes where terminals not only convey data to the other terminals but also state-information that they extract from their previous observations. This state-information can be exploited at the other terminals to improve their sensing performances. Indeed, as we show through examples, our schemes improve over previous non-collaborative schemes in terms of their achievable rate-distortion tradeoffs. For D2D we propose two schemes, one where compression of state information is separated from channel coding and one where it is integrated via a hybrid coding approach (10.1109/JSAIT.2023.3286932)
  DOI : 10.1109/JSAIT.2023.3286932
• Towards Efficient, General and Robust Entity Disambiguation Systems
  
  • Chen Lihu
  , 2023. Entity disambiguation aims to map mentions in documents to standard entities in a given knowledge base, which is important for various applications such as information extraction, Web search and question answering. Although the field is very vibrant with many novel works popping up, there are three questions that are underexplored by prior work. 1) Can we use a small model to approach the performance of a big model? 2) How to develop a single disambiguation system adapted to multiple domains? 3) Are existing systems robust to out-of-vocabulary words and different word orderings? Based on the three questions, we explore how to construct an efficient, general and robust entity disambiguation system. We also successfully apply entity disambiguation to the knowledge base completion task, especially for the long-tail entities