Constructive Synthesis of Memory-Intensive Accelerators for FPGA From Nested Loop Kernels

Field-programmable gate arrays are ideal hosts to custom accelerators for signal, image, and data processing but de- mand manual register transfer level design if high performance and low cost are desired. High-level synthesis reduces this design burden but requires manual design of complex on-chip and off-chip memory architectures, a major limitation in applications such as video processing. This paper presents an approach to resolve this shortcoming. A constructive process is described that can derive such accelerators, including on- and off-chip memory storage from a C description such that a user-defined throughput constraint is met. By employing a novel statement-oriented approach, dataflow intermediate models are derived and used to support simple ap- proaches for on-/off-chip buffer partitioning, derivation of custom on-chip memory hierarchies and architecture transformation to ensure user-defined throughput constraints are met with minimum cost. When applied to accelerators for full search motion estima- tion, matrix multiplication, Sobel edge detection, and fast Fourier transform, it is shown how real-time performance up to an order of magnitude in advance of existing commercial HLS tools is enabled whilst including all requisite memory infrastructure. Further, op- timizations are presented that reduce the on-chip buffer capacity and physical resource cost by up to 96% and 75%, respectively, whilst maintaining real-time performance.

Simulation of Distributed Contact in String Instruments: a Modal Expansion Approach

Impactive contact between a vibrating string and a barrier is a strongly nonlinear phenomenon that presents several challenges in the design of numerical models for simulation and sound synthesis of musical string instruments. These are addressed here by applying Hamiltonian methods to incorporate distributed contact forces into a modal framework for discrete-time simulation of the dynamics of a stiff, damped string. The resulting algorithms have spectral accuracy, are unconditionally stable, and require solving a multivariate nonlinear equation that is guaranteed to have a unique solution. Exemplifying results are presented and discussed in terms of accuracy, convergence, and spurious high-frequency oscillations.

FPGA-based soft-core processors for image processing applications

With security and surveillance, there is an increasing need to process image data efficiently and effectively either at source or in a large data network. Whilst a Field-Programmable Gate Array (FPGA) has been seen as a key technology for enabling this, the design process has been viewed as problematic in terms of the time and effort needed for implementation and verification. The work here proposes a different approach of using optimized FPGA-based soft-core processors which allows the user to exploit the task and data level parallelism to achieve the quality of dedicated FPGA implementations whilst reducing design time. The paper also reports some preliminary
progress on the design flow to program the structure. An implementation for a Histogram of Gradients algorithm is also reported which shows that a performance of 328 fps can be achieved with this design approach, whilst avoiding the long design time, verification and debugging steps associated with conventional FPGA implementations.

Wavelength converters based on four-wave mixing in semiconductor optical amplifiers for advanced modulation formats

Les convertisseurs de longueur d’onde sont essentiels pour la réalisation de réseaux de communications optiques à routage en longueur d’onde. Dans la littérature, les convertisseurs de longueur d’onde basés sur le mélange à quatre ondes dans les amplificateurs optiques à semi-conducteur constituent une solution extrêmement intéressante, et ce, en raison de leurs nombreuses caractéristiques nécessaires à l’implémentation de tels réseaux de communications. Avec l’émergence des systèmes commerciaux de détection cohérente, ainsi qu’avec les récentes avancées dans le domaine du traitement de signal numérique, il est impératif d’évaluer la performance des convertisseurs de longueur d’onde, et ce, dans le contexte des formats de modulation avancés. Les objectifs de cette thèse sont : 1) d’étudier la faisabilité des convertisseurs de longueur d’onde basés sur le mélange à quatre ondes dans les amplificateurs optiques à semi-conducteur pour les formats de modulation avancés et 2) de proposer une technique basée sur le traitement de signal numérique afin d’améliorer leur performance. En premier lieu, une étude expérimentale de la conversion de longueur d’onde de formats de modulation d’amplitude en quadrature (quadrature amplitude modulation - QAM) est réalisée. En particulier, la conversion de longueur d’onde de signaux 16-QAM à 16 Gbaud et 64-QAM à 5 Gbaud dans un amplificateur optique à semi-conducteur commercial est réalisée sur toute la bande C. Les résultats démontrent qu’en raison des distorsions non-linéaires induites sur le signal converti, le point d’opération optimal du convertisseur de longueur d’onde est différent de celui obtenu lors de la conversion de longueur d’onde de formats de modulation en intensité. En effet, dans le contexte des formats de modulation avancés, c’est le compromis entre la puissance du signal converti et les non-linéarités induites qui détermine le point d’opération optimal du convertisseur de longueur d’onde. Les récepteurs cohérents permettent l’utilisation de techniques de traitement de signal numérique afin de compenser la détérioration du signal transmis suite à sa détection. Afin de mettre à profit les nouvelles possibilités offertes par le traitement de signal numérique, une technique numérique de post-compensation des distorsions induites sur le signal converti, basée sur une analyse petit-signal des équations gouvernant la dynamique du gain à l’intérieur des amplificateurs optiques à semi-conducteur, est développée. L’efficacité de cette technique est démontrée à l’aide de simulations numériques et de mesures expérimentales de conversion de longueur d’onde de signaux 16-QAM à 10 Gbaud et 64-QAM à 5 Gbaud. Cette méthode permet d’améliorer de façon significative les performances du convertisseur de longueur d’onde, et ce, principalement pour les formats de modulation avancés d’ordre supérieur tel que 64-QAM. Finalement, une étude expérimentale exhaustive de la technique de post-compensation des distorsions induites sur le signal converti est effectuée pour des signaux 64-QAM. Les résultats démontrent que, même en présence d’un signal à bruité à l’entrée du convertisseur de longueur d’onde, la technique proposée améliore toujours la qualité du signal reçu. De plus, une étude du point d’opération optimal du convertisseur de longueur d’onde est effectuée et démontre que celui-ci varie en fonction des pertes optiques suivant la conversion de longueur d’onde. Dans un réseau de communication optique à routage en longueur d’onde, le signal est susceptible de passer par plusieurs étages de conversion de longueur d’onde. Pour cette raison, l’efficacité de la technique de post-compensation est démontrée, et ce pour la première fois dans la littérature, pour deux étages successifs de conversion de longueur d’onde de signaux 64-QAM à 5 Gbaud. Les résultats de cette thèse montrent que les convertisseurs de longueur d’ondes basés sur le mélange à quatre ondes dans les amplificateurs optiques à semi-conducteur, utilisés en conjonction avec des techniques de traitement de signal numérique, constituent une technologie extrêmement prometteuse pour les réseaux de communications optiques modernes à routage en longueur d’onde.

Tensor Completion for Multidimensional Inverse Problems with Applications to Magnetic Resonance Relaxometry

This thesis deals with tensor completion for the solution of multidimensional inverse problems. We study the problem of reconstructing an approximately low rank tensor from a small number of noisy linear measurements. New recovery guarantees, numerical algorithms, non-uniform sampling strategies, and parameter selection algorithms are developed. We derive a fixed point continuation algorithm for tensor completion and prove its convergence. A restricted isometry property (RIP) based tensor recovery guarantee is proved. Probabilistic recovery guarantees are obtained for sub-Gaussian measurement operators and for measurements obtained by non-uniform sampling from a Parseval tight frame. We show how tensor completion can be used to solve multidimensional inverse problems arising in NMR relaxometry. Algorithms are developed for regularization parameter selection, including accelerated k-fold cross-validation and generalized cross-validation. These methods are validated on experimental and simulated data. We also derive condition number estimates for nonnegative least squares problems. Tensor recovery promises to significantly accelerate N-dimensional NMR relaxometry and related experiments, enabling previously impractical experiments. Our methods could also be applied to other inverse problems arising in machine learning, image processing, signal processing, computer vision, and other fields.

Sparse Signal Representation in Digital and Biological Systems

Theories of sparse signal representation, wherein a signal is decomposed as the sum of a small number of constituent elements, play increasing roles in both mathematical signal processing and neuroscience. This happens despite the differences between signal models in the two domains. After reviewing preliminary material on sparse signal models, I use work on compressed sensing for the electron tomography of biological structures as a target for exploring the efficacy of sparse signal reconstruction in a challenging application domain. My research in this area addresses a topic of keen interest to the biological microscopy community, and has resulted in the development of tomographic reconstruction software which is competitive with the state of the art in its field. Moving from the linear signal domain into the nonlinear dynamics of neural encoding, I explain the sparse coding hypothesis in neuroscience and its relationship with olfaction in locusts. I implement a numerical ODE model of the activity of neural populations responsible for sparse odor coding in locusts as part of a project involving offset spiking in the Kenyon cells. I also explain the validation procedures we have devised to help assess the model's similarity to the biology. The thesis concludes with the development of a new, simplified model of locust olfactory network activity, which seeks with some success to explain statistical properties of the sparse coding processes carried out in the network.

Exploitation du potentiel sismique des étoiles naines blanches

Le but de cette thèse est d’explorer le potentiel sismique des étoiles naines blanches pulsantes, et en particulier celles à atmosphères riches en hydrogène, les étoiles ZZ Ceti. La technique d’astérosismologie exploite l’information contenue dans les modes normaux de vibration qui peuvent être excités lors de phases particulières de l’évolution d’une étoile. Ces modes modulent le flux émergent de l’étoile pulsante et se manifestent principalement en termes de variations lumineuses multi-périodiques. L’astérosismologie consiste donc à examiner la luminosité d’étoiles pulsantes en fonction du temps, afin d’en extraire les périodes, les amplitudes apparentes, ainsi que les phases relatives des modes de pulsation détectés, en utilisant des méthodes standards de traitement de signal, telles que des techniques de Fourier. L’étape suivante consiste à comparer les périodes de pulsation observées avec des périodes générées par un modèle stellaire en cherchant l’accord optimal avec un modèle physique reconstituant le plus fidèlement possible l’étoile pulsante. Afin d’assurer une recherche optimale dans l’espace des paramètres, il est nécessaire d’avoir de bons modèles physiques, un algorithme d’optimisation de comparaison de périodes efficace, et une puissance de calcul considérable. Les périodes des modes de pulsation de modèles stellaires de naines blanches peuvent être généralement calculées de manière précise et fiable sur la base de la théorie linéaire des pulsations stellaires dans sa version adiabatique. Afin de définir dans son ensemble un modèle statique de naine blanche propre à l’analyse astérosismologique, il est nécessaire de spécifier la gravité de surface, la température effective, ainsi que différents paramètres décrivant la disposition en couche de l’enveloppe. En utilisant parallèlement les informations obtenues de manière indépendante (température effective et gravité de surface) par la méthode spectroscopique, il devient possible de vérifier la validité de la solution obtenue et de restreindre de manière remarquable l’espace des paramètres. L’exercice astérosismologique, s’il est réussi, mène donc à la détermination précise des paramètres de la structure globale de l’étoile pulsante et fournit de l’information unique sur sa structure interne et l’état de sa phase évolutive. On présente dans cette thèse l’analyse complète réussie, de l’extraction des fréquences à la solution sismique, de quatre étoiles naines blanches pulsantes. Il a été possible de déterminer les paramètres structuraux de ces étoiles et de les comparer remarquablement à toutes les contraintes indépendantes disponibles dans la littérature, mais aussi d’inférer sur la dynamique interne et de reconstruire le profil de rotation interne. Dans un premier temps, on analyse le duo d’étoiles ZZ Ceti, GD 165 et Ross 548, afin de comprendre les différences entre leurs propriétés de pulsation, malgré le fait qu’elles soient des étoiles similaires en tout point, spectroscopiquement parlant. L’analyse sismique révèle des structures internes différentes, et dévoile la sensibilité de certains modes de pulsation à la composition interne du noyau de l’étoile. Afin de palier à cette sensibilité, nouvellement découverte, et de rivaliser avec les données de qualité exceptionnelle que nous fournissent les missions spatiales Kepler et Kepler2, on développe une nouvelle paramétrisation des profils chimiques dans le coeur, et on valide la robustesse de notre technique et de nos modèles par de nombreux tests. Avec en main la nouvelle paramétrisation du noyau, on décroche enfin le ”Saint Graal” de l’astérosismologie, en étant capable de reproduire pour la première fois les périodes observées à la précision des observations, dans le cas de l’étude sismique des étoiles KIC 08626021 et de GD 1212.

A study on independent component analysis over galois fields

Dissertação (mestrado)—Universidade de Brasília, Faculdade de Tecnologia, Departamento de Engenharia Elétrica, 2015.

Sensor inteligente para medição de cargas mecânicas

A medição precisa da força é necessária para muitas aplicações, nomeadamente, para a determinação da resistência mecânica dos materiais, controlo de qualidade durante a produção, pesagem e segurança de pessoas. Dada a grande necessidade de medição de forças, têm-se desenvolvido, ao longo do tempo, várias técnicas e instrumentos para esse fim. Entre os vários instrumentos utilizados, destacam-se os sensores de força, também designadas por células de carga, pela sua simplicidade, precisão e versatilidade. O exemplo mais comum é baseado em extensómetros elétricos do tipo resistivo, que aliados a uma estrutura formam uma célula de carga. Este tipo de sensores possui sensibilidades baixas e em repouso, presença de offset diferente de zero, o que torna complexo o seu condicionamento de sinal. Este trabalho apresenta uma solução para o condicionamento e aquisição de dados para células de carga que, tanto quanto foi investigado, é inovador. Este dispositivo permite efetuar o condicionamento de sinal, digitalização e comunicação numa estrutura atómica. A ideia vai de encontro ao paradigma dos sensores inteligentes onde um único dispositivo eletrónico, associado a uma célula de carga, executa um conjunto de operações de processamento de sinal e transmissão de dados. Em particular permite a criação de uma rede ad-hoc utilizando o protocolo de comunicação IIC. O sistema é destinado a ser introduzido numa plataforma de carga, desenvolvida na Escola Superior de Tecnologia e Gestão de Bragança, local destinado à sua implementação. Devido à sua estratégia de conceção para a leitura de forças em três eixos, contém quatro células de carga, com duas saídas cada, totalizando oito saídas. O hardware para condicionamento de sinal já existente é analógico, e necessita de uma placa de dimensões consideráveis por cada saída. Do ponto de vista funcional, apresenta vários problemas, nomeadamente o ajuste de ganho e offset ser feito manualmente, tornando-se essencial um circuito com melhor desempenho no que respeita a lidar com um array de sensores deste tipo.

Processing of laser Doppler flowmetry signals from healthy subjects and patients with varicose veins: information categorization approach based on intrinsic mode functions and entropy computation

International audience

Structural health monitoring of storage tanks

In this thesis, Ph.D candidate presents a compact sensor node (SN) designed for long-term and real-time acoustic emission (AE) monitoring of above ground storage tanks (ASTs). Each SN exploits up to three inexpensive low-frequency sensors based on piezoelectric diaphragms for effective leakage detection, and it is capable by means of built-in Digital Signal Processing functionalities to process the acquired time waveforms extracting the AE features usually required by testing protocols. Alternatively, capability to plug three high frequency AE sensors to a SN for corrosion simulated phenomena detection is envisaged and demonstrated. Another innovative aspect that the Ph.D candidate presents in this work is an alternative mathematical model of corrosion location on the bottom of the AST. This approach implies considering the three-dimensional localization model versus the two-dimensional commonly used according to the literature. This approach is aimed at significant optimization in the number of sensors in relation to the standard approach for solving localization problems as well as to allow filtering the false AE events related to the condensate droplets from AST ceiling. The technological implementation of this concept required the solution of a number of technical problems, such as the precise time of arrival (ToA) signal estimation, vertical localization of the AE source and multilaration solution that were discussed in detail in this work. To validate the developed prototype, several experimental campaigns were organized that included the simulation of target phenomena both in laboratory conditions and on a real water storage tank. The presented test results demonstrate the successful application of the developed AE system both for simulated leaks and for corrosion processes on the tank bottom. Mathematical and technological algorithms for localization and characterization of AE signals implemented during the development of the prototype are also confirmed by the test results.

Wireless solutions for the industrial internet of things

The fourth industrial revolution is paving the way for Industrial Internet of Things applications where industrial assets (e.g., robotic arms, valves, pistons) are equipped with a large number of wireless devices (i.e., microcontroller boards that embed sensors and actuators) to enable a plethora of new applications, such as analytics, diagnostics, monitoring, as well as supervisory, and safety control use-cases. Nevertheless, current wireless technologies, such as Wi-Fi, Bluetooth, and even private 5G networks, cannot fulfill all the requirements set up by the Industry 4.0 paradigm, thus opening up new 6G-oriented research trends, such as the use of THz frequencies. In light of the above, this thesis provides (i) a broad overview of the main use-cases, requirements, and key enabling wireless technologies foreseen by the fourth industrial revolution, and (ii) proposes innovative contributions, both theoretical and empirical, to enhance the performance of current and future wireless technologies at different levels of the protocol stack. In particular, at the physical layer, signal processing techniques are being exploited to analyze two multiplexing schemes, namely Affine Frequency Division Multiplexing and Orthogonal Chirp Division Multiplexing, which seem promising for high-frequency wireless communications. At the medium access layer, three protocols for intra-machine communications are proposed, where one is based on LoRa at 2.4 GHz and the others work in the THz band. Different scheduling algorithms for private industrial 5G networks are compared, and two main proposals are described, i.e., a decentralized scheme that leverages machine learning techniques to better address aperiodic traffic patterns, and a centralized contention-based design that serves a federated learning industrial application. Results are provided in terms of numerical evaluations, simulation results, and real-world experiments. Several improvements over the state-of-the-art were obtained, and the description of up-and-running testbeds demonstrates the feasibility of some of the theoretical concepts when considering a real industry plant.

UNIQUENESS AND NONUNIQUENESS RESULTS FOR A CERTAIN CLASS OF ALMOST PERIODIC DISTRIBUTIONS

We consider distributions u is an element of S'(R) of the form u(t) = Sigma(n is an element of N) a(n)e(i lambda nt), where (a(n))(n is an element of N) subset of C and Lambda = (lambda n)(n is an element of N) subset of R have the following properties: (a(n))(n is an element of N) is an element of s', that is, there is a q is an element of N such that (n(-q) a(n))(n is an element of N) is an element of l(1); for the real sequence., there are n(0) is an element of N, C > 0, and alpha > 0 such that n >= n(0) double right arrow vertical bar lambda(n)vertical bar >= Cn(alpha). Let I(epsilon) subset of R be an interval of length epsilon. We prove that for given Lambda, (1) if Lambda = O(n(alpha)) with alpha < 1, then there exists epsilon > 0 such that u vertical bar I(epsilon) = 0 double right arrow u 0; (2) if Lambda = O(n) is uniformly discrete, then there exists epsilon > 0 such that u vertical bar I(epsilon) = 0 double right arrow u 0; (3) if alpha > 1 and. is uniformly discrete, then for all epsilon > 0, u vertical bar I(epsilon) = 0 double right arrow u = 0. Since distributions of the above mentioned form are very common in engineering, as in the case of the modeling of ocean waves, signal processing, and vibrations of beams, plates, and shells, those uniqueness and nonuniqueness results have important consequences for identification problems in the applied sciences. We show an identification method and close this article with a simple example to show that the recovery of geometrical imperfections in a cylindrical shell is possible from a measurement of its dynamics.

Determination of Three-Dimensional Left Ventricle Motion to Analyze Ventricular Dyssyncrony in SPECT Images

A method to compute three-dimension (3D) left ventricle (LV) motion and its color coded visualization scheme for the qualitative analysis in SPECT images is proposed. It is used to investigate some aspects of Cardiac Resynchronization Therapy (CRT). The method was applied to 3D gated-SPECT images sets from normal subjects and patients with severe Idiopathic Heart Failure, before and after CRT. Color coded visualization maps representing the LV regional motion showed significant difference between patients and normal subjects. Moreover, they indicated a difference between the two groups. Numerical results of regional mean values representing the intensity and direction of movement in radial direction are presented. A difference of one order of magnitude in the intensity of the movement on patients in relation to the normal subjects was observed. Quantitative and qualitative parameters gave good indications of potential application of the technique to diagnosis and follow up of patients submitted to CRT.