In this dissertation, the operation of dual-linear polarized antennas is considered in order to provide ideal performance suited for several applications including polarimetric synthetic aperture radar (SAR), wireless and satellite communications. The underlying objectives realized in this work are reported as design realizations of dual-linear polarized antennas with low cross polarization patterns and high isolation between ports that employ special properties of the electromagnetic metamaterial (MTM) structures. Some of these key properties appear as negative permittivity, negative permeability, negative refractive index, and antiparallel nature of the phase velocity and the group velocity. The antenna design is carried out at two frequencies, 5.5 GHz and 10 GHz, and key physical issues that affect the operation of dual-linear polarization operation antennas are treated in light of electromagnetic MTM properties. It’s well known that a dual linear polarized antenna poses a big challenges such as cross polarization patterns and high mutual coupling between two input ports. Therefore, these drawbacks are key topic that receive significant attention in literature which reports on how to mitigate these drawbacks, however, at the expense of complexity of the antenna structures. The MTM structures have received considerable coverage in antenna research for obtaining size reduction, directivity enhancement, and beam steering. For this purpose, different MTMs structures are chosen in this thesis for achieving additional improvements, while keeping the antenna design as simple as possible, something which is very difficult to accomplish using conventional design methods. / In der folgenden Dissertation wird der Einsatz von zweifach linear polarisierten Antennen zur idealen Ausführung von verschiedenen Anwendungen, einschließlich von polarimetrischen Synthetic Aperture Radar (SAR), kabellose und satellitengestützte Kommunikation, diskutiert. Die Ziele dieser Arbeit werden dargestellt durch die Gestaltung von zweifach linear polarisierten Antennen mit gering Kreuz-Polarisationsmustern und die starke Isolation zwischen den Ports durch die einzigartigen Eigenschaften der Strukturen des elektromagnetischen Metamaterials (electromagnetic metamaterial; MTM). Einige dieser Eigenschaften treten als negative Permittivität, negative Permeabilität, negativer Brechungsindex und als antiparallel Richtungen (Gegenvektor) der Phasen-und Gruppengeschwindigkeit auf. Somit wird die Antennengestaltung auf zwei Frequenzen übertragen, 5,5GHz und 10 GHz, und die Ausführung der zweifach linearen Polarisation wird durch die elektromagnetischen Eigenschaften des MTM illustriert. Weil die Kreuzpolarisationsmuster und starke gegenseitige Koppelung zwischen zwei Input-Ports bei einer zweifach linear polarisierten Antenne große Schwierigkeiten bereiten, werden diese im Großteil der Fachliteratur als Schwerpunkte gesetzt, was zu einer Milderung der Nachteile führte, jedoch dafür die Komplexität der Antennenstruktur zunahm. Die Vielfalt an MTM ist ein bedeutender Teil im Bereich der Antennenforschung einschließlich der Größenverkleinerung, der Verbesserung der Richtcharakteristik und der Strahlensteuerung. Für diesen Zweck werden in dieser Dissertation verschiedenste MTM Strukturen ausgewählt um weitere Verbesserungen der Antennenstruktur zu ermöglichen und gleichzeitig die Einfachheit der Struktur zu bewahren, was mit konventionellen Gestaltungsmethoden nur schwer zu erreichen ist.

The computer performance has been improved tremendously since the development of the first allpurpose, all electronic digital computer in 1946. However, engineers, scientists and researchers keep making more efforts to further improve the computer performance to meet the demanding requirements for many applications. There are basically two ways to improve the computer performance in terms of computational speed. One way is to use faster devices (VLSI chips). Although faster and faster VLSI components have contributed a great deal on the improvement of computation speed, the breakthroughs in increasing switching speed and circuit densities of VLSI devices will be diflicult and costly in future. The other way is to use parallel processing architectures which employ multiple processors to perform a computation task. When multiple processors working together, an appropriate architecture is very important to achieve the maximum performance in a cost-effective manner. Systolic arrays are ideally qualified for computationally intensive applications with inherent massive parallelism because they capitalize on regular, modular, rhythmic, synchronous, concurrent processes that require intensive, repetitive computation. This thesis can be divided into three parts. The first part is an introductory part containing Chap. I and Chap. 2. The second part, composed of Chap. 3 and Chap. 4 concerns with the systolic design methodology. The third part deals with the several systolic array design for different applications.

Noroviruses are the major cause of acute viral gastroenteritis worldwide. Thus, rapid and reliable pathogen detection and control are crucial to avoid epidemic outbreaks. Peptides which bind to these viruses with high specificity and affinity could serve as small and stable recognition elements in biosensing applications for a point-of-care diagnostic of noroviruses. They can be identified by screening large phage display libraries using the biopanning technique. In the present study, this method was applied to identify norovirus-binding peptide motifs. For this purpose, a biopanning based on column chromatography was established, and three rounds of selections were performed. After the second round, the cosmix-plexing recombination technique was implemented to enhance the chance of obtaining peptides with very high affinity. Biopanning data evaluation was based on next-generation sequencing (NGS), to show that this innovative method can enable a detailed analysis of the complete sequence spectrum obtained during and after biopanning. Highly enriched motifs could be characterized by their large proportion of the amino acids W, K, R, N, and F. Neighbourhood analysis was exemplarily performed for selected motifs, showing that the motifs FAT, RWN, and KWF possessed the fingerprints with the largest differences relative to the original library. This thesis thus presents next-generation sequencing-based analysis tools, which could now be transferred to any other biopanning project. The identified peptide motifs represent promising candidates for a future examination of their norovirus-specific binding. A new option for testing such phage-target interactions in the context of biopanning selections was studied in the second part of the thesis. For this purpose, a phage-based microarray was developed as a miniaturized binding assay. As a prerequisite, the different immobilization behaviour of phages on positively and negatively charged surfaces was studied, and a non-contact printing technique for bacteriophages was developed. Subsequently, the interaction of phages and antibodies directed against phage coat proteins was characterized in enzyme-linked immunosorbent assays, and the protocol was successfully transferred to the non-contact printed phage spots. At the proof-of-concept level, the phage array could finally be integrated into a microfluidic setup, showing a higher signal-to-background ratio relative to the static phage array. These results point the way towards a microfluidic phage array, allowing online monitoring, automation, and parallelisation of the phage array analysis. / Noroviren gelten als Hauptursache akuter viraler Magen-Darm-Erkrankungen. Nur eine zeitnahe und verlässliche Detektion und Kontrolle dieser Pathogene kann epidemische Ausbrüche vermeiden. Um dies zu ermöglichen, könnten Peptide, die an diese Viren mit hoher Spezifität und Affinität binden, als kleine und stabile Erkennungselemente in biosensorischen Anwendungen eingesetzt werden. Solche Peptide können mithilfe der Biopanning-Technik identifiziert werden, die auf dem Screening großer Phagen-Display-Bibliotheken beruht. In der vorliegenden Arbeit wurde diese Methode genutzt, um Norovirus-bindende Peptidmotive zu identifizieren. Dazu wurde ein auf Säulenchromatographie basierendes Biopanning entwickelt und drei Selektionsrunden durchgeführt. Die Cosmix-Plexing-Rekombinationstechnik wurde nach der zweiten Runde eingesetzt, um die Wahrscheinlichkeit der Gewinnung hochaffiner Binder zu erhöhen. Die Auswertung der Biopanningdaten erfolgte mittels Hochdurchsatzsequenzierung (Next-Generation Sequencing). Es konnte gezeigt werden, dass diese innovative Methode die detailierte Analyse des kompletten Sequenzspektrums während und nach dem Biopanning ermöglicht. Stark angereicherte Motive konnten durch ihren hohen Anteil an den Aminosäuren W, K, R, N und F charakterisiert werden. Eine Nachbarschaftsanalyse wurde exemplarisch für ausgewählte Motive durchgeführt. Dabei wurden die stärksten Unterschiede im Fingerprint im Vergleich zur Ausgangsbibliothek bei den Motiven FAT, RWN und KWF gefunden. Diese Dissertation stellt damit auf Next-Generation Sequencing basierende Analysetechniken vor, die für weitere Biopanningprojekte übernommen werden können. Die identifizierten Peptidmotive könnten als vielversprechende Kandidaten zukünftig auf ihre Norovirus-spezifische Bindung hin getestet werden. Eine neue Möglichkeit, solche Phagen-Analyt-Interaktionen zu untersuchen, wurde im zweiten Teil der Dissertation untersucht. Dafür wurde als miniaturisierter Bindungsassay ein Phagen-basiertes Mikroarray entwickelt. Als Voraussetzung wurde zunächst das unterschiedliche Immobilisierungsverhalten von Bakteriophagen auf positiv und negativ geladenen Oberflächen untersucht und eine kontaktfreie Drucktechnik für Bakteriophagen etabliert. Anschließend wurde die Interaktion von Phagen und gegen sie gerichteten Antikörpern in Enzym-gekoppelten Immunadsorptionstests charakterisiert und das Protokoll erfolgreich auf die kontaktfrei gedruckten Phagenspots übertragen. Schließlich wurde erstmals die grundsätzliche Möglichkeit gezeigt, das Array in ein mikrofluidisches Setup zu integrieren, was zu einem höheren Signal-zu-Hintergrund-Verhältnis im Vergleich zum statischen Array führte. Diese Ergebnisse zeigen damit den Weg zu einem mikrofluidischen Phagen-Array auf, das sowohl die Möglichkeit des Online-Monitorings als auch der Automatisierung und Parallelisierung der Phagen-Array-Analyse bietet.

Recently, covering arrays have been the subject of considerable research attention as they hold both theoretical interest and practical importance due to their applications to testing. In this thesis, we perform the first comprehensive study of a generalization of covering arrays called variable strength covering arrays, where we dictate the interactions to be covered in the array by modeling them as facets of an abstract simplicial complex. We outline the necessary background in the theory of hypergraphs, combinatorial testing, and design theory that is relevant to the study of variable strength covering arrays. We then approach questions that arise in variable strength covering arrays in a number of ways. We demonstrate their connections to hypergraph homomorphisms, and explore the properties of a particular family of abstract simplicial complexes, the qualitative independence hypergraphs. These hypergraphs are tightly linked to variable strength covering arrays, and we determine and identify several of their important properties and subhypergraphs. We give a detailed study of constructions for variable strength covering arrays, and provide several operations and divide-and-conquer techniques that can be used in building them. In addition, we give a construction using linear feedback shift registers from primitive polynomials of degree 3 over arbitrary finite fields to find variable strength covering arrays, which we extend to strength-3 covering arrays whose sizes are smaller than many of the best known sizes of covering arrays. We then give an algorithm for creating variable strength covering arrays over arbitrary abstract simplicial complexes, which builds the arrays one row at a time, using a density concept to guarantee that the size of the resultant array is asymptotic in the logarithm of the number of facets in the abstact simplicial complex. This algorithm is of immediate practical importance, as it can be used to create test suites for combinatorial testing. Finally, we use the Lovasz Local Lemma to nonconstructively determine upper bounds on the sizes of arrays for a number of different families of hypergraphs. We lay out a framework that can be used for many hypergraphs, and then discuss possible strategies that can be taken in asymmetric problems.

Placement is one of the most important steps in physical design for VLSI circuits. For field programmable gate arrays (FPGAs), the placement step determines the location of each logic block. I present novel timing and congestion driven placement algorithms for FPGAs with minimal runtime overhead. By predicting the post-routing timing-critical edges and estimating congestion accurately, this algorithm is able to simultaneously reduce the critical path delay and the minimum number of routing tracks. The core of the algorithm consists of a criticality-history record of connection edges and a congestion map. This approach is applied to the 20 largest Microelectronics Center of North Carolina (MCNC) benchmark circuits. Experimental results show that compared with the state-of-the-art FPGA place and route package, the Versatile Place and Route (VPR) suite, this algorithm yields an average of 8.1% reduction (maximum 30.5%) in the critical path delay and 5% reduction in channel width. Meanwhile, the average runtime of the algorithm is only 2.3X as of VPR.

A galectina-1 (Gal-1) é uma lectina homodimérica multifuncional capaz de reconhecer e se ligar a beta-galactosídeos por meio de um domínio denominado carbohydrate recognition domain (CRD). A Gal-1 humana (Gal-1h) e a Gal-1 de camundongo (Gal-1c) mantêm 88,15% de homologia e, apesar de não existirem mutações em aminoácidos-chave do CRD, há substituições próximas a esses resíduos. Considerando as implicações dessas diferenças em estrutura e função, e que é comum a utilização de modelos murinos para estudar a função Gal-1, o presente trabalho objetiva analisar comparativamente a Gal-1c e a Gal-1h por meio de ensaios de cristalização e determinação estrutural da Gal-1c, além da avaliação comparativa da atividade lectínica da Gal-1h e da Gal-1c por glycan array e hemaglutinação. Também foi avaliada a capacidade de ambas as Gal-1 em induzir a exposição de fosfatidilserina (FS) em neutrófilos ativados provenientes de medula de camundongos normais ou deficientes de ?-2 integrina (Mac-1), de modo a investigar se a interação Gal-1/Mac-1 estaria envolvida nesse processo. Preparações homogêneas e ativas de Gal-1c e Gal-1h foram utilizadas nos ensaios. Os cristais de Gal-1c foram obtidos em 20% de polietilenoglicol 3350 e 0,2 M de fluoreto de amônio. Os dados de difração de raios X foram coletados e processados, obtendo-se uma estrutura com resolução de 2,4 Å. Observou-se que substituições de aminoácidos entre a Gal-1c e a Gal-1h estão localizadas em regiões expostas ao solvente, próximas do CRD e distantes da interface de dimerização. A análise comparativa entre Gal-1c e Gal-1h mostrou que estas substituições conferem a Gal-1c um caráter mais polar, com consequente aumento da distribuição de volume molecular. Nos ensaios de hemaglutinação, pode-se observar que é necessária uma concentração 2 vezes maior de Gal-1c para aglutinar eritrócitos humanos, de carneiro e de coelho na mesma proporção que a Gal-1h. Por meio do glycan array, pode-se determinar o perfil de ligação a glicanas de ambas as Gal-1. As duas Gal-1 apresentam afinidade por glicanas ramificadas contendo galactose terminal, e a Gal-1h apresentou maior intensidade de ligação às glicanas quando comparada à Gal-1c. Preparações de Gal-1c e Gal-1h induzem níveis semelhantes de exposição de FS na superfície de neutrófilos deficientes ou não de Mac-1, sugerindo que a interação Gal-1/Mac-1 não esteja envolvida no processo de exposição de FS na superfície de neutrófilos ativados. Assim, a diferença sequencial entre a Gal-1c e a Gal-1h é capaz de gerar diferenças estruturais consideráveis que implicam no reconhecimento diferencial de glicanas, o que, entretanto, não se reflete na capacidade de indução de FS na superfície de neutrófilos ativados. Além disso, a interação Gal-1/Mac-1 parece não participar desse processo, o que pode indicar que o papel da Gal-1 no turnover de neutrófilos, via reconhecimento fagocítico, seja um processo complexo e independente dessa interação. / Galectin-1 (Gal-1) is a homodimeric and multifunctional lectin that recognizes and binds to beta-galactoside by a carbohydrate recognition domain (CRD). Human Gal-1 (hGal-1) and mouse Gal-1 (mGal-1) are 88.15% identical, and although there are no mutations in key amino acids within the CRD, there are differences in the amino acids sequence near the CRD. Given the potential of these differences to alter overall structure and function, and the common utilization of murine models to study Gal-1 function, we sought to directly compare key biochemical features of hGal and mGal-1. Thus, we performed crystallization and structure determination assays of mGal-1, and determined the carbohydrate binding specificy of mGal-1 and hGal-1 using a glycan array and using hemagglutination assay. We also evaluated the ability of both Gal-1 to induce exposure of phosphatidylserine (PS) in activated neutrophils from the bone marrow of normal or ?-2 integrin (Mac-1) deficient mice, in order to investigate the involvement of Gal-1/Mac-1 interaction in this process. To accomplish this, homogeneous and active preparations of hGal-1 and mGal-1 were used in the study. mGal-1 crystals were obtained in 20% polyethylene glycol 3350 and 0.2 M ammonium fluoride. Data from X-ray diffraction were collected and processed, yielding a structure with a final resolution of 2.4 Å. The amino acid substitutions found between mGal-1 and hGaI-1 are detected on the solvent-exposed surfaces where the CRDs are located and not on the proteins dimerization surfaces. A comparative structural analysis between mGal-1 and hGal-1 shows that these amino acid substitutions confer to mGal-1 a greater number of ionizable residues, polar character, appearance of the acid regions clustered, and a slight increase of volume distribution. In hemagglutination assays, twice the concentration of mGal-1 was required to cause equivalent agglutination of human, sheep or rabbit erythrocytes as hGal-1. Glycan array analysis demonstrated that both galectins have affinity for branched glycans containing terminal galactose residues. However, hGal-1 appeared to display higher levels of binding that mGal-1. Preparations of mGal-1 and hGal-1 induced similar levels of PS exposure on normal or Mac-1 deficient neutrophils, suggesting that the interaction Gal-1/Mac-1 is not involved in this process. Thus, hGal-1 and mGal-1 appear to possess considerable differences in glycan recognition that likely reflects subtle difference in amino acid sequence. Furthermore, the interaction Gal-1/Mac-1 do not appear to participate in this PS exposure process, which suggest that other Gal-1 receptors are likely important in this process.

Ce travail s'intéresse au suivi spatio-temporel par imagerie ultrasonore de la cavitation acoustique. Celle-ci est un phénomène physique complexe utilisé au cours de certaines techniques de thérapie par ultrasons, correspondant à la formation de bulles de gaz qui oscillent et éclatent. Initialement, la méthode TD-PAM (Time Domain Passive Acoustic Mapping, en anglais), a été développée pour cartographier l’activité de cavitation à partir des signaux acoustiques émis par les bulles, enregistrés passivement par une sonde linéaire d'imagerie ultrasonore. Toutefois, le TD-PAM souffre d’une trop faible résolution et de nombreux artefacts de reconstruction. De plus, il est lourd en temps de calcul car il est formalisé dans le domaine temporel (TD). Pour pallier ces deux limitations, il est proposé d'étudier, de comparer et de développer des méthodes avancées d'imagerie ultrasonore passive. Ce manuscrit s'articule autour de trois contributions principales : Une méthode adaptative originale a été formalisée dans le domaine temporel, reposant sur la compression d'amplitude des signaux ultrasonores par racine pième : le TD-pPAM. Cette approche améliore la résolution et le contraste des cartes de cavitation pour un temps de calcul équivalent au TD-PAM. La notion de matrice de densité inter-spectrale a été introduite pour l'imagerie de la cavitation. Dès lors, quatre méthodes dans le domaine de Fourier (FD) ont été étudiées et comparées : le FD-PAM (non-adaptatif), la méthode Robuste de Capon FD-RCB (adaptatif, par optimisation), le Functional Beamforming FD-FB (adaptatif, par compression non-linéaire) et la méthode MUltiple Signal Classification FD-MUSIC (adaptatif, par projection en sous-espaces). Les performances de ces méthodes FD ont été étudiées expérimentalement in vitro cuve d’eau avec une comparaison par imagerie optique. Les méthodes adaptatives FD proposées ont démontré leur potentiel à améliorer le suivi spatio-temporel des bulles. Le FD-RCB offre une localisation supérieure au FD-PAM mais souffre d'une importante complexité algorithmique. Les performances du FD-FB sont intermédiaires à celles du FD-PAM et du FD-RCB, pour une complexité de calcul équivalente au FD-PAM. Le FD-MUSIC a le potentiel de mettre en évidence de faibles sources acoustiques, mais ne conserve pas leurs quantifications relatives / This work focuses on the spatio-temporal monitoring of acoustic cavitation by ultrasonic imaging. This is a complex physical phenomenon used in some ultrasound therapy techniques, corresponding to the formation of gas bubbles that oscillate and implode. Initially, the TD-PAM (Time Domain Passive Acoustic Mapping) method was developed to map cavitation activity from acoustic signals emitted by bubbles, passively recorded by a linear ultrasonic imaging probe. However, the TD-PAM suffers from too low resolution and many reconstruction artifacts. In addition, it is time-consuming because it is formalized in the time domain (TD). To overcome these two limitations, it is proposed to study, compare and develop advanced methods of passive ultrasound imaging. This manuscript is structured around three main contributions: An original adaptive method has been formalised in the time domain, based on the amplitude compression of ultrasonic signals by root pth: TD-pPAM. This approach improves the resolution and contrast of cavitation maps for a computing time equivalent to the TD-PAM. The notion of cross-spectral density matrix has been introduced for cavitation imaging. Four Fourier domain (FD) methods were therefore studied and compared: FD-PAM (non-adaptive), Capon Robuste FD-RCB (adaptive, by optimization), Functional Beamforming FD-FB (adaptive, by non-linear compression) and MUltiple Signal Classification FD-MUSIC (adaptive, by subspaces projection). The performance of these FD methods was studied experimentally in vitro in water tank with a comparison by optical imaging. The proposed adaptive FD methods have demonstrated their potential to improve the spatial and temporal tracking of bubbles. The FD-RCB offers a superior localization to the FD-PAM but suffers from a high algorithmic complexity. The performance of the FD-FB is intermediate to that of the FD-PAM and the FD-RCB, for a calculation complexity equivalent to the FD-PAM. The FD-MUSIC has the potential to highlight weak acoustic sources, but does not keep their relative quantifications

For a linear array, the excitation coefficients of each element and its geometry play an important role, because they will determine the radiation pattern of the given array. Side Lobe Level (SLL) is one of the key parameters to evaluate the radiation pattern of the array. Generally speaking, we desire SLL to be as low as possible. For the linear array with uniform spacing, there are some classic methods to calculate the excitation coefficients to make the radiation pattern satisfy the given requirements. For the linear array with non-uniform spacing, linear programming and particle swarm optimization are proposed to calculate the excitation coefficients to make the array get minimum SLL in this thesis. They are demonstrated for symmetric and asymmetric array in the first part of this thesis. In the second part of this thesis, a simple method is proposed for correcting excitation coefficients of a linear phased array. This proposed method corrects the coefficients through using the Normalized Least Means Squares(NLMS) algorithm, dither signal and a near-field sensor being used for sensing the field emitted by the array. The advantage of this proposed method is that it avoids the problem of estimating the largest eigenvalue of the coefficient matrix to get optimal step size. Its robustness in different environments is demonstrated as well as the effect of noise with various Signal-to-Noise Ratio (SNR), and mutual coupling. In addition, the effect of using discrete dither signal to the array is considered, because the continuous dither signal cannot be generated in practice.

Radio astronomy, the survey and study of naturally occurring astronomical radio signals, is a challenging field in terms of engineering requirements. The typical astronomical signal of interest is incredibly faint, resulting in very low signal-to-noise ratios (SNRs) on the order of -30 dB or lower. To detect such signals, one must have an uncommonly low and stable receiver noise temperature, high gain through large aperture reflectors, and state-of-the-art signal processing algorithms. One must also be able to mitigate the effects of interference, the presence of which, even if extremely weak, can completely mask the faint astronomical signals of interest. To this end, this work presents the development of and results from a new broadband phased array feed (PAF) named the Focal L-Band Array for the Green Bank Telescope (FLAG). This instrument is able to form multiple simultaneous beams to survey a large patch of sky instantaneously, and has a minimum system noise temperature (Tsys) of 16.83 K. This PAF also has the potential to use spatial filtering techniques to place pattern nulls in the direction of interfering signals through the use of an orthogonal projection. This work will also present an improved method for computing an orthogonal projection operator, which is able to place a spatially broad null in the direction of a moving RFI source. A formal derivation of some detection and estimation theory properties for astronomical radio transients is also presented, which formalization is lacking within the astronomical community. This includes maximum-likelihood detectors and estimators and a Cramér Rao bound (CRB) analysis of astronomical transient parameters.

De nos jours, la complexité de la conception des circuits intégrés et du logiciel croit régulièrement, faisant croître le besoin de la vérification dans chaque étape du cycle de conception. Le prototypage matériel sur une plateforme multi-FPGA présente le meilleur compromis entre le temps de conception d'un circuit et le temps d'exécution d'une application par ce circuit. Pour l'implémenter sur cette plateforme, une opération de partitionnement est effectuée avant de créer des partitions capables de s'intégrer dans chaque FPGAPar conséquent, des signaux coupés à l'interface des partitions doivent passer d'un FPGA à un autre. Cependant, le nombre de traces physiques inter-FPGA est limité ce qui crée des problèmes de routabilité du circuit prototypé. Cette thèse touche surtout la partie post-partitionnement et s'intéresse au problème deroutage inter-FPGA. Ainsi, les principaux travaux de cette thèse sont les suivants :Dans un premier temps, nous nous intéressons au développement d'un générateur debenchmarks qui permet, à l'aide d'une description architecturale simple du benchmark, de générer un circuit modélisé avec le langage de description matérielle VHDL. Le générateur utilise un ensemble de composants ce qui donne aux benchmarks un aspect réel semblable à celui des circuits industriels. Ces circuits de tests nous serviront pour évalue rles performances des techniques développées dans cette thèse. Dans un deuxième temps, nous proposons de développer un outil spécifique qui intervient après le partitionnement pour prendre en compte la contrainte liée à la limitation du nombre d'interconnexion entre les FPGAs. Cet outil est basé sur une approcheitérative visant à réduire le taux de multiplexage (nombre de signaux qui partagent un seul _l physique). Le routage en lui même est assuré par l'algorithme de routage Pathfinder adapté. Cet algorithme servira comme point de départ pour les techniques de routage développées durant cette thèse. Des adaptations adéquates seront faites pour cibler un ré-seau de routage inter-FPGA. Dans une deuxième partie, nous essayons de déterminer la meilleure forme du signal à router (bi-points ou multi-points) ainsi que le graphe de routage utilisé. Pour cela, nous proposons des scénarios de test a_n de sélectionner les critères qui donnent la fréquence de fonctionnement la plus performante. Par la suite, nous présentons une description détaillée des IPs de multiplexage utilisés.Ces IPs sont insérés dans les parties émettrices et réceptrices d'un canal de communication. Ces IPs incluent des composants spécifiques appelés SERDES pour assurer la sérialisation/déserialisation des données à transmettre. L'insertion de ces composants peut créer des problèmes de routabilité intra-FPGA. Ainsi, dans une deuxième partie, nous proposons un algorithme de placement basé sur l'estimation de la congestion afin d'améliorer la routabilité du circuit. / This thesis mainly deals with the post-partitioning task and addresses the problem of inter-FPGA routing. Thus, the main contributions of this thesis are: Firstly, we focus on the development of a benchmark generator which, using a simple architectural description of the benchmark, generates a circuit modelled with the hardware description language VHDL. The generator uses a set of industrial components providing benchmarks with real behaviour similar to that of industrial circuits. These benchmarks are used to evaluate the performance of the techniques developed in this thesis. In a second step , we propose a speci_c tool which acts after the partitioning to handle the constraints related to the limited number of interconnection between FPGAs. This tool is based on an iterative approach and aims to reduce the multiplexing ratio (the number of signals that share the same physical wire). The routing task itself is operated by the Pathfinder routing algorithm which is widely used by academic and industrial researchers . This algorithm is used as a starting point for routing techniques developed in this thesis . In a second part , we try to identify the best shape of the routed signals and the appropriate routing graph. For this reason, we propose scenarios to select criteria that give the best system frequency. Finally, we present a detailed description of the architecture of the multiplexing IPs. These IPs are inserted in the transmitting and receiving FPGAs of a communication channel. These IPs include speci_c components called SERDES for serialization/deserialization of the data. The insertion of these IPs can create problems of intra-FPGA routability. Thus, in a second part, we propose a placement algorithm based on congestion estimation to improve the routability of the circuit.