Publications

2022

• Generic Constructions of (Boolean and Vectorial) Bent Functions and Their Consequences
  
  • Li Yanjun
  • Kan Haibin
  • Mesnager Sihem
  • Peng Jie
  • Tan Chik How
  • Zheng Lijing
  IEEE Transactions on Information Theory, Institute of Electrical and Electronics Engineers, 2022, 68 (4), pp.2735-2751. (10.1109/TIT.2022.3140180)
  DOI : 10.1109/TIT.2022.3140180
• Writing-only in-MRAM computing paradigm for ultra-low power applications
  
  • Liu Bo
  • Liu Mingyue
  • Zhou Yongliang
  • Hong Xiaofeng
  • Cai Hao
  • Naviner Lirida
  Microprocessors and Microsystems: Embedded Hardware Design, Elsevier, 2022, 90, pp.104449. In-memory computing (IMC) is demonstrated as a breakthrough in terms of few interconnection delay and high density. Computing in nonvolatile memory, e.g., spin-transfer-torque (STT)-MRAM, is expected to realize near-zero leakage power consumption and unlimited endurance. Based on circuit-level reconfiguration, we propose a in-MRAM computing scheme to perform data storage and Boolean arithmetic simultaneously. This proposed scheme is implemented with a 28-nm CMOS process and a 40-nm Magnetic tunnel junction (MTJ) compact model. The self-write-termination method achieves 84.7% writing energy saving within 20-ns write access duration. A high-level simulation is performed with pixels image similarity analysis. It demonstrates 24% dynamic energy reduction comparing to the previous method. The proposed in-MRAM computing scheme is also applicable to other resistive memories with the identical bit-cell structure. (10.1016/j.micpro.2022.104449)
  DOI : 10.1016/j.micpro.2022.104449
• Agreement in Spiking Neural Networks
  
  • Kunev Martin
  • Kuznetsov Petr
  • Sheynikhovich Denis
  Journal of Computational Biology, Mary Ann Liebert, 2022, 29 (4), pp.358 - 369. We study the problem of binary agreement in a spiking neural network (SNN). We show that binary agreement on n inputs can be achieved with O(n) of auxiliary neurons. Our simulation results suggest that agreement can be achieved in our network in O(log n) time. We then describe a subclass of SNNs with a biologically plausible property, which we call size-independence. We prove that solving a class of problems, including agreement and Winner-Take-All, in this model requires O(n) auxiliary neurons, which makes our agreement network size-optimal. (10.1089/cmb.2021.0365)
  DOI : 10.1089/cmb.2021.0365
• Matrix Factorization for Blind Beam Alignment in Massive mmWave MIMO
  
  • Ktari Aymen
  • Ghauch Hadi
  • Rekaya Ben Othman Ghaya
  , 2022.
• Convergence of constant step stochastic gradient descent for non-smooth non-convex functions
  
  • Bianchi Pascal
  • Hachem Walid
  • Schechtman Sholom
  Set-Valued and Variational Analysis, Springer, 2022, 30 (3), pp.1117-1147. This paper studies the asymptotic behavior of the constant step Stochastic Gradient Descent for the minimization of an unknown function F , defined as the expectation of a non convex, non smooth, locally Lipschitz random function. As the gradient may not exist, it is replaced by a certain operator: a reasonable choice is to use an element of the Clarke subdifferential of the random function; an other choice is the output of the celebrated backpropagation algorithm, which is popular amongst practionners, and whose properties have recently been studied by Bolte and Pauwels [7]. Since the expectation of the chosen operator is not in general an element of the Clarke subdifferential BF of the mean function, it has been assumed in the literature that an oracle of BF is available. As a first result, it is shown in this paper that such an oracle is not needed for almost all initialization points of the algorithm. Next, in the small step size regime, it is shown that the interpolated trajectory of the algorithm converges in probability (in the compact convergence sense) towards the set of solutions of the differential inclusion. Finally, viewing the iterates as a Markov chain whose transition kernel is indexed by the step size, it is shown that the invariant distribution of the kernel converge weakly to the set of invariant distribution of this differential inclusion as the step size tends to zero. These results show that when the step size is small, with large probability, the iterates eventually lie in a neighborhood of the critical points of the mean function F . (10.1007/s11228-022-00638-z)
  DOI : 10.1007/s11228-022-00638-z
• Lower Bound on the Capacity of the Continuous-Space SSFM Model of Optical Fiber
  
  • Sefidgaran Milad
  • Yousefi Mansoor
  IEEE Transactions on Information Theory, Institute of Electrical and Electronics Engineers, 2022, 68 (4), pp.2460-2478. The capacity of a discrete-time model of optical fiber described by the split-step Fourier method (SSFM) as a function of the signal-to-noise ratio SNR and the number of segments in distance $K$ is considered. It is shown that if $K\geq \text {SNR} ^{2/3}$ and $\text {SNR} \rightarrow \infty $ , the capacity of the resulting continuous-space lossless model is lower bounded by $\frac {1}{2}\log _{2}(1+ \text {SNR}) - \frac {1}{2}+ o(1)$ , where $o(1)$ tends to zero with SNR. As $K \rightarrow \infty $ , the inter-symbol interference (ISI) averages out to zero due to the law of large numbers and the SSFM model tends to a diagonal phase noise model. It follows that, in contrast to the discrete-space model where there is only one signal degree-of-freedom (DoF) at high powers, the number of DoFs in the continuous-space model is at least half of the input dimension $n$ . Intensity-modulation and direct detection achieves this rate. The pre-log in the lower bound when $K= \sqrt [\delta]{ \text {SNR}}$ is generally characterized in terms of $\delta $ . It is shown that if the nonlinearity parameter $\gamma \rightarrow \infty $ , the capacity of the continuous-space model is $\frac {1}{2}\log _{2}(1+ \text {SNR})+ o(1)$ . The SSFM model when the dispersion matrix does not depend on $K$ is considered. It is shown that the capacity of this model when $K= \sqrt [\delta]{ \text {SNR}}$ , $\delta >3$ , and $\text {SNR} \rightarrow \infty $ is $\frac {1}{2n}\log _{2}(1+ \text {SNR})+ O(1)$ . Thus, there is only one DoF in this model. Finally, it is found that the maximum achievable information rates (AIRs) of the SSFM model with back-propagation equalization obtained using numerical simulation follows a double-ascent curve. The AIR characteristically increases with SNR, reaching a peak at a certain optimal power, and then decreases as SNR is further increased. The peak is attributed to a balance between noise and stochastic ISI. However, if the power is further increased, the AIR will increase again, approaching the lower bound $\frac {1}{2}\log (1+ \text {SNR})- \frac {1}{2} + o(1)$ . The second ascent is because the ISI averages out to zero with $K \rightarrow \infty $ sufficiently fast. (10.1109/TIT.2021.3139179)
  DOI : 10.1109/TIT.2021.3139179
• Influence des mises à jour de sécurité sur le comportement d'un système critique
  
  • Jaillon Philippe
  • Apvrille Ludovic
  , 2022. Les systèmes cyber-physiques sont des systèmes critiques qui sont capable d’interagir avec leur environnement par le biais de capteurs et d’actionneurs. Ces éléments étant contrôlés par des logiciels, une cyber-attaque qui les toucherait aurait des conséquences inacceptables. Il convient donc de les protéger des attaques en déployant des contre-mesures, tout en s’assurant que les modifications apportées n’induisent pas à leur tour des comportements néfastes aux systèmes. Dans cet exposé, nous présenterons W-Sec, une méthode conçue par Mines de St-Eienne et Télécom Paris dans le cadre du projet européen SPARTA. Cette méthode vise à évaluer quantitativement, en amont de leur déploiement, les impacts que peuvent avoir des contremesures sur des systèmes cyber-physiques. Cette évaluation, qui se fait en termes de sûreté, de sécurité et de performances et s’appuie sur l’utilisation de TTool, un outil de modélisation et de vérification formelles développé au sein de l’équipe LabSoC de Télécom Paris.
• Multimode Physics in the Mode Locking of Semiconductor Quantum Dot Lasers
  
  • Grillot Frédéric
  • Chow Weng W
  • Dong Bozhang
  • Ding Shihao
  • Huang Heming
  • Bowers John E
  Applied Sciences, Multidisciplinary digital publishing institute (MDPI), 2022, 12 (7), pp.1-14. Quantum dot lasers are an attractive option for light sources in silicon photonic integrated circuits. Thanks to the three-dimensional charge carrier confinement in quantum dots, high material gain, low noise and large temperature stability can be achieved. This paper discusses, both theoretically and experimentally, the advantages of silicon-based quantum dot lasers for passive mode-locking applications. Using a frequency domain approach, i.e., with the laser electric field described in terms of a superposition of passive cavity eigenmodes, a precise quantitative description of the conditions for frequency comb and pulse train formation is supported, along with a concise explanation of the progression to mode locking via Adler’s equation. The path to transform-limited performance is discussed and compared to the experimental beat-note spectrum and mode-locked pulse generation. A theory/experiment comparison is also used to extract the experimental group velocity dispersion, which is a key obstacle to transform-limited performance. Finally, the linewidth enhancement contribution to the group velocity dispersion is investigated. For passively mode-locked quantum dot lasers directly grown on silicon, our experimental and theoretical investigations provide a self-consistent accounting of the multimode interactions giving rise to the locking mechanism, gain saturation, mode competition and carrier-induced refractive index. (10.3390/app12073504)
  DOI : 10.3390/app12073504
• A Deep Residual Learning Implementation of Metamorphosis
  
  • Maillard Matthis
  • François Anton
  • Glaunès Joan
  • Bloch Isabelle
  • Gori Pietro
  , 2022. In medical imaging, most of the image registration methods implicitly assume a one-to-one correspondence between the source and target images (i.e., diffeomorphism). However, this is not necessarily the case when dealing with pathological medical images (e.g., presence of a tumor, lesion, etc.). To cope with this issue, the Metamorphosis model has been proposed. It modifies both the shape and the appearance of an image to deal with the geometrical and topological differences. However, the high computational time and load have hampered its applications so far. Here, we propose a deep residual learning implementation of Metamorphosis that drastically reduces the computational time at inference. Furthermore, we also show that the proposed framework can easily integrate prior knowledge of the localization of topological changes (e.g., segmentation masks) that can act as spatial regularization to correctly disentangle appearance and shape changes. We test our method on the BraTS 2021 dataset, showing that it outperforms current state-of-the-art methods in the alignment of images with brain tumors.
• Statistical Depth Functions for Ranking Distributions: Definitions, Statistical Learning and Applications
  
  • Goibert Morgane
  • Clémençon Stéphan
  • Irurozki Ekhine
  • Mozharovskyi Pavlo
  , 2022. The concept of median/consensus has been widely investigated in order to provide a statistical summary of ranking data, i.e. realizations of a random permutation Σ of a finite set, {1,. .. , n} with n ≥ 1 say. As it sheds light onto only one aspect of Σ's distribution P , it may neglect other informative features. It is the purpose of this paper to define analogues of quantiles, ranks and statistical procedures based on such quantities for the analysis of ranking data by means of a metric-based notion of depth function on the symmetric group. Overcoming the absence of vector space structure on S n , the latter defines a center-outward ordering of the permutations in the support of P and extends the classic metric-based formulation of consensus ranking (medians corresponding then to the deepest permutations). The axiomatic properties that ranking depths should ideally possess are listed, while computational and generalization issues are studied at length. Beyond the theoretical analysis carried out, the relevance of the novel concepts and methods introduced for a wide variety of statistical tasks are also supported by numerous numerical experiments.
• Analysis of a Target-Based Actor-Critic Algorithm with Linear Function Approximation
  
  • Barakat Anas
  • Bianchi Pascal
  • Lehmann Julien
  , 2022, 151. Actor-critic methods integrating target networks have exhibited a stupendous empirical success in deep reinforcement learning. However, a theoretical understanding of the use of target networks in actor-critic methods is largely missing in the literature. In this paper, we reduce this gap between theory and practice by proposing the first theoretical analysis of an online target-based actor-critic algorithm with linear function approximation in the discounted reward setting. Our algorithm uses three different timescales: one for the actor and two for the critic. Instead of using the standard single timescale temporal difference (TD) learning algorithm as a critic, we use a two timescales target-based version of TD learning closely inspired from practical actor-critic algorithms implementing target networks. First, we establish asymptotic convergence results for both the critic and the actor under Markovian sampling. Then, we provide a finite-time analysis showing the impact of incorporating a target network into actor-critic methods.
• Attack Graph-based Solution for Vulnerabilities Impact Assessment in Dynamic Environment
  
  • Boudermine Antoine
  • Khatoun Rida
  • Choyer Jean-Henri
  , 2022, pp.24-31. (10.1109/CIoT53061.2022.9766588)
  DOI : 10.1109/CIoT53061.2022.9766588
• Real-Time Locomotion on Soft Grounds With Dynamic Footprints
  
  • Alvarado Eduardo
  • Paliard Chloé
  • Rohmer Damien
  • Cani Marie-Paule
  Frontiers in Virtual Reality, Frontiers, 2022, 3, pp.801856: 1-12. When we move on snow, sand, or mud, the ground deforms under our feet, immediately affecting our gait. We propose a physically based model for computing such interactions in real time, from only the kinematic motion of a virtual character. The force applied by each foot on the ground during contact is estimated from the weight of the character, its current balance, the foot speed at the time of contact, and the nature of the ground. We rely on a standard stress-strain relationship to compute the dynamic deformation of the soil under this force, where the amount of compression and lateral displacement of material are, respectively, parameterized by the soil’s Young modulus and Poisson ratio. The resulting footprint is efficiently applied to the terrain through procedural deformations of refined terrain patches, while the addition of a simple controller on top of a kinematic character enables capturing the effect of ground deformation on the character’s gait. As our results show, the resulting footprints greatly improve visual realism, while ground compression results in consistent changes in the character’s motion. Readily applicable to any locomotion gait and soft soil material, our real-time model is ideal for enhancing the visual realism of outdoor scenes in video games and virtual reality applications. (10.3389/frvir.2022.801856)
  DOI : 10.3389/frvir.2022.801856
• Improving The Automatic Segmentation Of Elongated Organs Using Geometrical Priors
  
  • Vétil Rebeca
  • Bône Alexandre
  • Vullierme Marie-Pierre
  • Rohé Marc-Michel
  • Gori Pietro
  • Bloch Isabelle
  , 2022. Deep neural networks are widely used for automated organ segmentation as they achieve promising results for clinical applications. Some organs are more challenging to delineate than others, for instance due to low contrast at their boundaries. In this paper, we propose to improve the segmentation of elongated organs thanks to Geometrical Priors that can be introduced during training, using a local Tversky loss function, or at post-processing, using local thresholds. Both strategies do not introduce additional training parameters and can be easily applied to any existing network. The proposed method is evaluated on the challenging problem of pancreas segmentation. Results show that Geometrical Priors allow us to correct the systematic under-segmentation pattern of a state-of-the-art method, while preserving the overall segmentation quality.
• A Statistical Assessment of Anthropomorphic Characteristics Impacts on WBAN Communications
  
  • Youssef Badre
  • Roblin Christophe
  , 2022, pp.1-5. In Wireless Body Area Networks (WBAN), the subject characteristics (morphology, including gender, posture and movement) are among the main sources of the channel variability. The objective of this communication is to highlight statistically the influence of the anthropomorphic characteristics on the on-body channel. To do it, we have simulated five radio links on five homogeneous men subjects with different morphological characteristics. Experimental results obtained for some radio links have been used to support the relevance of the simulations. Within the framework of this first semi-quantitative approach, this communication is limited to the modeling of the Path Loss as a function of two quantitative morphological parameters that are the Body Mass Index (considering different body heights and fatness) and the Waist Circumference. The objective is to identify the main trends and to extract "semi-quantitative" models giving the orders of magnitude of the variability for each radio link. The latter typically ranges from 4 to 10.6 dB for the phantoms considered (except for a link for which the variations are limited to approximately 2 dB). This analysis is motivated by the importance of the variability obtained. (10.23919/EuCAP53622.2022.9769394)
  DOI : 10.23919/EuCAP53622.2022.9769394
• Indoor Material Transmission Measurements Between 2 GHz and 170 GHz for 6G Wireless Communication Systems
  
  • Aliouane Mohamed
  • Conrat Jean-Marc
  • Cousin Jean-Christophe
  • Begaud Xavier
  , 2022.
• Characterization of EMF Exposure in Massive MIMO Antenna Networks with Max-Min Fairness Power Control
  
  • Hajj Maarouf Al
  • Wang Shanshan
  • Wiart Joe
  , 2022, pp.1-5. In this paper, we analyze the EMF exposure, in terms of total received power, in the massive multiple-input multiple-output (MIMO) networks. With the recent deployment of 5G networks, the potential risks of electromagnetic field (EMF) exposure are gaining increasing attention. However, most of the current research that focus on the mathematical modeling of 5G networks ignore downlink power control. Therefore, we derive the framework of the average power received at nearest mobile terminal MT under max-min fairness power control using stochastic geometry. The total received power consists of three parts, useful signal, multi-user interference and inter-cell interference. We propose a tight approximation on the power control coefficient. Based on the proposed approximation, the framework on total received power is then validated by Monte-Carlo simulations. The results show that the average received power monotonically increases as the density of the base station increases and the number of users increases. (10.23919/EuCAP53622.2022.9768997)
  DOI : 10.23919/EuCAP53622.2022.9768997
• Path Loss Measurements and Modelling in a Citrus Plantation in the 1800 MHz, 3.5 GHz and 28 GHz in LoS
  
  • Juan-Llacer Leandro
  • Molina-Garcia-Pardo Jose Maria
  • Sibille Alain
  • Torrico Saul
  • Rubiola Luis Martinez
  • Martinez-Ingles Maria Teresa
  • Rodriguez Jose-Victor
  • Pascual-Garcia Juan
  , 2022, pp.1-5. Agriculture 4.0 is going to represent a massive deployment of sensors, so efficient planning of radiocommunication systems in this type of environment will be necessary. In this work, the measured path loss in a LoS condition, with the transmitter and receiver heights below the trees height, at a citrus plantation in the 1800 MHz, 3.5 GHz and 28 GHz frequency bands using a two-ray model in the vertical plane and the FI (Floating Intercept) and the CI (Close-In reference) slope models is analysed. Results show that, in the 28 GHz band, the direct and reflected-from-soil contribution may be enough to estimate the path loss for all the measure distance range, however, in the 1800 MHz and 3.5 GHz bands the multiple-scattering contributions from trees need to be considered after some distance between the transmitter and receiver. Furthermore, a guiding effect has been observed only in the 1800 MHz band. (10.23919/EuCAP53622.2022.9769016)
  DOI : 10.23919/EuCAP53622.2022.9769016
• EME-Net: A U-net-based Indoor EMF Exposure Map Reconstruction Method
  
  • Mallik Mohammed
  • Kharbech Sofiane
  • Mazloum Taghrid
  • Wang Shanshan
  • Wiart Joe
  • Gaillot Davy
  • Clavier Laurent
  , 2022, pp.1-5. In wireless communication systems, in order to respond to the perception of risks related to electromagnetic field exposure and allocate radio resources, the estimation of the received power and exposure map is an essential task and a challenge. This paper proposes an algorithm for estimating electromagnetic field exposure maps using U-net architecture based on convolutional neural networks. The power map estimation is transformed into an image reconstruction task by image color mapping, where every pixel value of the image represents received power intensity. The designed model learns wireless signal propagation characteristics in a realistic indoor environment while considering various positions of the Wi-Fi access points. Results show that indoor propagation phenomena and environment models can be learned from data producing an accurate power map to measure the electromagnetic field. (10.23919/EuCAP53622.2022.9769645)
  DOI : 10.23919/EuCAP53622.2022.9769645
• Réflecteur multibande à métamatériaux
  
  • Gonçalves Licursi de Mello Rafael
  • Begaud Xavier
  • Lepage Anne Claire
  , 2022. In this work, we present the design of a tri-band artificial magnetic conductor (AMC) as a quad-band reflector for the 5G, 4G, Wi-Fi and X-band communications. To show the interest of the solution, the AMC is placed under an Archimedean spiral at a distance close to a quarter-wavelength in the X band, which implies a smaller electrical distance for the 5G, 4G, and Wi-Fi. Compared to the spiral alone, an improvement in the broadside realized gain is obtained in all bands, with stable diagrams and a profile of 0.1 , where is the wavelength at 2.4 GHz.
• Laser à cavité externe accordable sur 174 nm intégrant une puce à boîtes quantiques
  
  • Ehlert Jannik F.
  • Mugnier Alain
  • Grillot Frederic
  , 2022.
• High speed mid-infrared Stark modulator for optical data transmission up to 10 Gbit.s-1
  
  • Bonazzi Thomas
  • Dely Hamza
  • Spitz Olivier
  • Rodriguez Etienne
  • Gacemi Djamal
  • Todorov Yanko
  • Pantzas Konstantinos
  • Beaudoin Grégoire
  • Sagnes Isabelle
  • Grillot Frédéric
  • Vasanelli Angela
  • Sirtori Carlo
  , 2022, pp.MF5C.4. Using the Stark effect in coupled InGas/AlInAs quantum wells, we demonstrate a mid-infrared broadband optoelectronic external modulator enabling 10 Gbit/s free space optical data-transmission in the second atmospheric window (9 µm) at room temperature. (10.1364/MICS.2022.MF5C.4)
  DOI : 10.1364/MICS.2022.MF5C.4
• Strategies to Extend the Bandwidth of Wideband Antennas Over Artificial Magnetic Conductors
  
  • Begaud Xavier
  , 2022. From the two last decades, there has been a growing interest in applying artificial materials, known as Metamaterials to antennas. These materials derive their unique properties, not from their composition but from their structure. They are mostly composed of a periodic arrangement of materials, patterns. This spatial periodicity naturally induces a spectral selectivity. This narrow bandwidth can be used with benefit but is usually cited, in addition to losses, as one of the main limitations for metamaterials applications. The objective of this presentation is to demonstrate that it is possible to design wideband antennas with metamaterials. We are going to focus our attention to those, which can be applied to improve performance in terms of gain and to reduce the thickness of the overall antenna. Designing unidirectional antennas is required on many platforms in order to obtain outward radiation and preserve the interior of any electromagnetic pollution. For this, most common solutions are to place the antenna above a reflector or cavity filled with absorbing materials. The solution with absorbing cavity is simple but half of the radiation is lost and the cavity is sized at a quarter of a wavelength at the lowest operating frequency, which become very bulky for low frequency applications. Another efficient technique is to use a reflector made of a very good electrical conductor to retrieve the radiation lost in the previous solution. This technique is optimal at the middle of the bandwidth where a constructive interference phenomenon occurs by placing the reflector at a quarter wavelength (at center frequency) from the antenna. But this solution is inherently limited bandwidth and can rarely exceed the octave. It therefore appears difficult to design a unidirectional antenna achieving wide bandwidth and good compactness. Nevertheless, reducing the thickness is possible by using an AMC (Artificial Magnetic Conductors) which exhibit the perfect magnetic conductor (PMC) behavior, i.e. no phase shift at reflexion. It then becomes possible to position the antenna closest to this new reflector. The antenna is so unidirectional and thin. However obtaining PMC condition is possible but in a very limited frequency band and generally the AMC has a limited size composed with only few cells who are not all illuminated by a plane wave at normal incidence … In this talk, several strategies to exceed these limitations and keeping the wideband performance of antennas placed above these artificial magnetic conductors will be detailed. The presented results will be compared with the measurements in order to validate the different designs.
• A benchmark of incremental model transformation tools based on an industrial case study with AADL
  
  • Mkaouar Hana
  • Blouin Dominique
  • Borde Etienne
  Software and Systems Modeling, Springer Verlag, 2022. (10.1007/s10270-022-00989-z)
  DOI : 10.1007/s10270-022-00989-z
• The bow‐tie antenna: Performance limitations and improvements
  
  • Gonçalves Licursi de Mello Rafael
  • Lepage Anne Claire
  • Begaud Xavier
  IET Microwaves Antennas and Propagation, Institution of Engineering and Technology, 2022. The simple planar shape and wideband input impedance of the bow-tie antenna make it suitable to diverse wideband applications. However, the radiation pattern of this kind of antenna is not stable and the gain in the broadside direction deteriorates with increasing frequencies. In this article, the operation of both the triangular and the rounded-edge versions of the bow-tie antenna is thoroughly investigated and the performance thereof is compared. A technique to maximise such performance, which consists of inserting radially aligned grooves in the radiators' ends, is presented. The bandwidth, defined by a broadside gain of at least 2 dBi and a reflection coefficient magnitude of less than −10 dB, is improved from 85.7% to 97.9% in simulations. Such an improvement is validated through the measurement of two prototypes, in which the presented technique also increases the broadside gain in up to 1.5 dB, with no significant change in the shapes of the radiation patterns and in the reflection coefficient magnitude. (10.1049/mia2.12242)
  DOI : 10.1049/mia2.12242