Most models of the musculoskeletal system incorporate individual or groups of muscles as a series of line segments, assuming all fiber bundles within a muscle have the same length and moment arm, and do not account for architectural differences throughout the muscle volume. The purpose was to develop an algorithm to fit digitized fiber bundle data from one specimen into muscle volume and intramuscular nerve distribution data from seven other specimens of extensor carpi radialis longus (ECRL) and brevis (ECRB). Coherent Point Drift (CPD) algorithm was successfully adapted for this purpose. The intramuscular nerve distribution and fiber bundle architecture was modeled in all the muscle volumes. ECRL was found to have two neuromuscular compartments, superficial and deep, while ECRB was found to have two, three or four, in a proximal to distal direction depending on the number of primary nerve branches.

Most models of the musculoskeletal system incorporate individual or groups of muscles as a series of line segments, assuming all fiber bundles within a muscle have the same length and moment arm, and do not account for architectural differences throughout the muscle volume. The purpose was to develop an algorithm to fit digitized fiber bundle data from one specimen into muscle volume and intramuscular nerve distribution data from seven other specimens of extensor carpi radialis longus (ECRL) and brevis (ECRB). Coherent Point Drift (CPD) algorithm was successfully adapted for this purpose. The intramuscular nerve distribution and fiber bundle architecture was modeled in all the muscle volumes. ECRL was found to have two neuromuscular compartments, superficial and deep, while ECRB was found to have two, three or four, in a proximal to distal direction depending on the number of primary nerve branches.

Master of Science / Department of Kinesiology / Thomas J. Barstow / Background and Aim: Long-term microgravity exposure, via spaceflight or -6° head-down tilt bedrest, has been shown to produce significant cardiovascular deconditioning and decreases in exercise performance. However, there is little known about how acute microgravity exposure influences the cardiovascular system’s ability to adjust to increases in physical work. Therefore, the aim of this study was to compare cardiovascular and exercise performance during acute upright, supine and -6° head-down tilt positions. Methods: Seven healthy inactive men performed maximal cycle exercise (VO2peak) tests in the upright, supine, and -6° head-down tilt on separate days. Oxygen consumption and heart rate were measured continuously throughout the testing procedures. Cardiac output (acetylene exhalation technique) was measured periodically and interpolated to the 100-watt work rate. Stroke volume was calculated from cardiac output and heart rate data. Results: Peak oxygen uptake and heart rate were significantly decreased in the supine and -6° head-down tilt positions compared to the upright (VO2peak 2.01±0.46, 2.01±0.51 versus 2.32±0.61 L/min respectively; peak heart rate 161±13, 160±14 versus 172±11 bmp). However, cardiac output at 100-watts was similar in all three-exercise positions. Calculated stroke volume at 100-watts was significantly higher in the -6° head-down tilt position compared to the upright position (76.6±4.7 versus 71.2±4.5, ml). Conclusion: These results suggest that exercise capacity is immediately decreased upon exposure to a microgravity environment, prior to any cardiovascular deconditioning. Therefore, an astronaut’s exercise performance should be evaluated with exercise tests in the -6° head-down tilt position prior to space flight in order to establish a baseline response.

The purpose of the study was to investigate the static and dynamic architecture of supraspinatus throughout its volume in the normal and pathological state. The architecture was first investigated in cadaveric specimens free of any tendon pathology. Using a serial dissection and digitization method tailored for supraspinatus, the musculotendinous architecture was modeled in situ. The 3D model reconstructed in Autodesk MayaTM allowed for visualization and quantification of the fiber bundle architecture i.e. fiber bundle length (FBL), pennation angle (PA), muscle volume (MV) and tendon dimensions. Based on attachment sites and architectural parameters, the supraspinatus was found to have two architecturally distinct regions, anterior and posterior, each with three subdivisions. The findings from the cadaveric investigation served as a map and platform for the development of an ultrasound (US) protocol that allowed for the dynamic fiber bundle architecture to be quantified in vivo in normal subjects and subjects with a full-thickness supraspinatus tendon tear. The architecture was studied in the relaxed state and in three contracted states (60º abduction with either neutral rotation, 80º external rotation, or 80º internal rotation). The dynamic changes in the architecture within the distinct regions of the muscle were not uniform and varied as a function of joint position. Mean FBL in the anterior region shortened significantly with contraction (p<0.05) but not in the posterior. In the anterior region, mean PA was significantly smaller in the middle part compared to the deep (p<0.05). Comparison of the normal and pathological muscle found large differences in the percentage change of FBL and PA with contraction. The architectural parameter that showed the largest changes with tendon pathology was PA. In sum, the results showed that the static and dynamic fiber bundle architecture of supraspinatus is heterogeneous throughout the muscle volume and may influence tendon stresses. The architectural data collected in this study and the 3D muscle model can be used to develop future contractile models. The US protocol may serve as an assessment tool to predict the functional outcome of rehabilitative exercises and surgery.

Η μερική αντικατάσταση των παραδοσιακών ενσύρματων επικοινωνιών από ασύρματες οδήγησε στην ανάγκη για κατάλληλη σχεδίαση των ασύρματων δικτύων, ώστε να υποστηρίζουν τα απαιτούμενα επίπεδα ποιότητας. Η προσφερόμενη ποιότητα περιορίζεται σίγουρα από φυσικά εμπόδια όπως τα όρια που έθεσε ο Shannon, αλλά και από την περιορισμένη γνώση μας στην εκπομπή και διάδοση κυμάτων. Η διαρκής βελτίωση της ικανότητας των ασύρματων δικτύων να προσφέρουν υψηλής ποιότητας υπηρεσίες είναι όμως γεγονός και οφείλεται σε δύο βασικούς ερευνητικούς άξονες, αυτόν που εφευρίσκει νέες μεθόδους επικοινωνίας (διαμόρφωση, κωδικοποίηση, πολυπλεξία, κεραίες κ.α.) και αυτόν που προσπαθεί να βελτιώσει τη διαχείριση των ήδη υπαρχόντων πόρων. Η παρούσα διατριβή προσφέρει επιστημονικές ιδέες και μοντέλα, που έχουν σαν σκοπό την καλύτερη διαχείριση των πόρων. Τα παραδοσιακά αλλά και τα μοντέρνα δίκτυα κινητής τηλεφωνίας έχουν κατακλύσει την αγορά. Η λειτουργία τους όμως, παραμένει να κινείται στα πλαίσια συμβιβασμών, προσφέροντας χαμηλή ποιότητα υπηρεσίας. Η είσοδος των ασύρματων δικτύων υπολογιστών στην αγορά, ως ανταγωνιστές παροχής κλασικών υπηρεσιών (υπηρεσία φωνής και υπηρεσία δεδομένων), δίνει μια ώθηση αναπροσαρμογής των στόχων των δικτύων κινητής τηλεφωνίας. Οι προτεινόμενοι αλγόριθμοι διαχείρισης των δικτύων αυτών προσφέρουν καλύτερη ποιότητα υπηρεσιών με αλλαγή μεμονωμένων μόνο τμημάτων του δικτύου, χωρίς την ασύμφορη ολική επανασχεδίασή του. Αποτελούν δηλαδή πρακτικές λύσεις για την αποδοτικότερη λειτουργία των ήδη εγκατεστημένων δικτύων. Από την άλλη πλευρά, η ανερχόμενη τεχνολογία των ασύρματων δικτύων υπολογιστών βρίσκεται περισσότερο σε ανταγωνισμό με τα κλασικά ενσύρματα δίκτυα υπολογιστών. Το πλεονέκτημα της ελεύθερης κίνησης μετατρέπεται σε δυσβάσταχτο περιορισμό όταν η σύγκριση ποιότητας των δύο δικτύων είναι αναπόφευκτη. Για το λόγο αυτό, υπάρχει η μεγάλη ανάγκη τα ασύρματα δίκτυα να γίνουν ανταγωνιστικά των ενσύρματων και κυρίως να παρέχουν ποιότητα ικανού επιπέδου, ώστε οι διάφορες υπηρεσίες να μπορούν να λειτουργήσουν ανεπηρέαστες από την χρήση του ασύρματου καναλιού. Οι προτεινόμενοι αλγόριθμοι σε αυτόν τον τομέα αποσκοπούν στην βελτίωση της λειτουργίας των ασύρματων δικτύων με τελικό σκοπό την εξασφάλιση επιπέδων ποιότητας. Για την μελέτη των ασύρματων δικτύων κινητής τηλεφωνίας και υπολογιστών και την εξαγωγή συμπερασμάτων απαιτούνται αναλυτικές μέθοδοι περιγραφής και εξομοιώσεις των δικτύων αυτών ώστε να γίνουν και οι κατάλληλες συγκρίσεις της προσφερόμενης ποιότητας. Ο τομέας αυτός της επιστήμης είναι διαρκώς ενεργός καθώς νέες τεχνικές και αναλύσεις προτείνονται συνεχώς. Με στόχο την ανάλυση και την εξομοίωση των προηγούμενα αναφερθέντων αλγορίθμων αναπτύχθηκε ένας αριθμός τεχνικών, που αποτελούν αναπόσπαστο τμήμα της διατριβής αυτής. Στα πλαίσια της διατριβής αυτής και συγκεκριμένα στο τμήμα μελέτης της επίδοσης των δικτύων, προτείνεται η μετατροπή μη γραμμικών αλυσίδων Markov σε γραμμικές, για χρήση σε προβλήματα κίνησης ταυτόχρονων πολλαπλών υπηρεσιών. Η προτεινόμενη μεθοδολογία οδηγεί σε ταχύτατο υπολογισμό της αλυσίδας με τη χρήση μιας προσέγγισης. Παράλληλα, αναλύεται ο χρόνος διαμονής και ο εναπομένων χρόνος διαμονής σε μια κυψέλη. Η ανάλυση γίνεται εκ του μηδενός και με χρήση βασικών στοχαστικών μοντέλων, ενώ το αποτέλεσμα καταλήγει σε κατανομή gamma, κάτι που είχε ήδη παρατηρηθεί από πειράματα στην βιβλιογραφία. Στη συνέχεια κατασκευάζεται ένα ντετερμινιστικό μοντέλο εξομοίωσης που μπορεί να εφαρμοστεί εύκολα μέσω ενός χάρτη σε κάθε πόλη με ορθογώνια δόμηση. Στο τμήμα των δικτύων κινητής τηλεφωνίας προτείνεται ένας αλγόριθμος διαχείρισης και αποδοχής κλήσεων που προκαλεί μεταπομπές συστήματος (system initiated handover). Η χρήση των μεταπομπών αυτών επιφέρει μεταβλητή χωρητικότητα (soft capacity) σε μια ομάδα κυψελών με αποτέλεσμα την αποδοτικότερη διαχείριση πόρων σε περιπτώσεις άνισης κατανομής κίνησης. Ένας άλλος αλγόριθμος προτείνεται για διαρκή διαπραγμάτευση (real-time negotiation) ποιότητας μεταξύ του δικτύου και του χρήστη με αποτέλεσμα την καλύτερη οργάνωση και διαχείριση των πόρων. Αποδεικνύονται η δυνατότητα χρήσης του αλγορίθμου σε πραγματικό χρόνο και τα οφέλη που αποκομίζονται από αυτήν. Στο τμήμα ασύρματων δικτύων υπολογιστών προτείνεται ένα νέο πρωτόκολλο πρόσβασης μέσου, για χρήση στα ασύρματα τοπικά δίκτυα. Το πρωτόκολλο αυτό είναι πλήρως κατανεμημένο (μπορεί να χρησιμοποιηθεί και σε αυτοοργανονούμενα δίκτυα - ad hoc), ενώ προσφέρει τη δυνατότητα παροχής διαφορετικών επιπέδων ποιότητας, από άριστη ποιότητα για κλήσεις πραγματικού χρόνου μέχρι ποιότητα καλύτερης δυνατής προσπάθειας (best effort). Επίσης προτείνονται αλγόριθμοι βελτίωσης των ασύρματων δικτύων αισθητήρων. Αναλύεται συγκεκριμένα ο αυτόματος εντοπισμός κόμβων με χρήση τριγωνισμού και μεταβλητής ισχύος σημάτων-φάρων και η εξοικονόμηση ενέργειας με χρήση συστημάτων πολλαπλών κεραιών (MIMO). / The trend of replacing the traditional wired communications with their wireless counterparts led to the need for better design and organization of wireless networks so as to provide the necessary Quality of Service (QoS) levels. The offered QoS is bounded by Shannon’s limit and from the limitation of human knowledge on the wireless channel matters. However, the constant improvement of network performance is a fact based on two parallel scientific axes, the one that provides new communication techniques (modulation, coding, multiple access, antennas, etc) and the one that handles the already invented methods in an optimized manner. This thesis offers new scientific ideas and models for better resource management of wireless networks. Traditional and modern mobile telephone networks have occupied the market. Nevertheless, their functionality is still based on compromising of offered quality with the needs of a two-way voice application. Wireless computer networks appear in the market as contenders of voice service, acting like a threat to telephone networks. The proposed models for the mobile telephone networks provide solutions for improving overall QoS by altering only minor parts of these already-installed networks. On the other hand, the advancement of the wireless technology gives rise to visions of the so called Broadband Wireless Access. Computer Networks are expected to offer global roaming and provide traditional services over the IP protocol. In this context, QoS is necessary for guaranteeing service levels of quality. The proposed models in this part are focused in offering acceptable quality levels over the wireless channel for these services. The performance analysis of wireless networks requires the use of analytical approaches and simulation of these networks in order to assess the final offered quality. This section of science is constantly developing since new methods and techniques are applied to different parts of research. Through the procedure of analyzing the wireless networks in this thesis, a number of new techniques have been invented which constitute an important part of it. In terms of network traffic description, a new transformation of nonlinear markov chain to linear is proposed. By means of this transformation, the calculation of unbalanced nonuniform traffic can be accelerated. In the section of performance analysis, the sojourn time and the remaining sojourn time are analytically calculated. These cell parameters are found to follow a gamma distribution, a result that is backed from simulations in bibliography. Moreover, a deterministic simulation model is derived to enable easy calculation of these parameters from a map. As regards the mobile telephone networks, a new call admission control scheme is proposed in order to simulate soft capacity functionality in 2G networks. This feature, organizes better the resources in a cluster in cases of inhomogeneous network traffic. Moreover, a QoS negotiation algorithm is proposed for use in 3G networks (e.g. UMTS). Real-time negotiation is proposed for better resource organization and management. The gain from these schemes is calculated and presented. A new Medium Access Control (MAC) protocol is proposed for use in 802.11 Wireless Local Area Networks. This protocol is backward compatible with the protocol 802.11e and offers better quality infrastructure for VoIP services. The proposed protocol is analyzed and simulated and the results are compared to the previously used protocols to showcase the improvements. A new analytical approach with better accuracy is proposed for this reason. Finally, two techniques are proposed for improving the quality of wireless sensor networks. Firstly, a technique for low cost real-time localization with a small number of GPS or other beacons is proposed and tested. Secondly, a means of energy conservatiοn by use of Multiple Input Multiple Output (MIMO) systems is proposed, and then the gain from a QoS scheme is derived.

Muscle architecture, innervation pattern and fiber type distribution of lumbar multifidus (LMT) throughout its volume was quantified. Musculotendinous (n=10) and neural components (n=3) were dissected and digitized from thirteen embalmed cadaveric specimens. The data were imported into Autodesk® Maya® 2008 to generate 3D neuromuscular models of each specimen. Architectural parameters (fiber bundle length, FBL; fiber bundle angle, FBA; tendon length) were quantified from the models using customized software. The medial branch of the posterior rami (L1-L5) was traced through LMT to determine its distribution. Using immunohistochemistry, Type I/II muscle fibers were identified in 29 muscle biopsies from one fresh frozen specimen. The total area and number of each cell type was calculated using Visiopharm® (image analysis software). Architectural and fiber type data were analyzed using ANOVA with Tukey’s post-hoc test (p ≤ 0.05). From L1-L4, LMT had three architecturally distinct regions: superficial, intermediate and deep. At L5, intermediate LMT was absent. Mean FBL decreased significantly from superficial (5.8 ± 1.6cm) to deep regions (2.9 ± 1.1cm) as did volume (superficial, 5.6 ± 2.3ml; deep, 0.7 ± 0.3ml). In contrast, mean FBA increased from superficial to deep. The medial branch of the posterior ramus (L1-L5) supplied the five bands of LMT. Each medial branch in turn divided to supply the deep, intermediate and superficial regions separately. The area occupied by Type I fibers was significantly less (p< 0.01) in the deep (56%) compared with the superficial regions (75%). Based on architecture and morphology, superficial LMT with the longest FBL and relatively small FBA is well designed for torque production and controlling the lumbar lordosis. Intermediate LMT with significantly longer FBL compared with the deep region and with its caudal to cranial line of action may help to control intersegmental stability. Furthermore, the absence of intermediate LMT at L5 and may contribute to the higher incidence of instability observed at the lumbosacral junction. Deep LMT with its short FBL, large FBA and proximity to the axis of spinal rotation may function to provide proprioceptive input to the CNS rather than a primary stabilizer of the lumbar spine.

Muscle architecture, innervation pattern and fiber type distribution of lumbar multifidus (LMT) throughout its volume was quantified. Musculotendinous (n=10) and neural components (n=3) were dissected and digitized from thirteen embalmed cadaveric specimens. The data were imported into Autodesk® Maya® 2008 to generate 3D neuromuscular models of each specimen. Architectural parameters (fiber bundle length, FBL; fiber bundle angle, FBA; tendon length) were quantified from the models using customized software. The medial branch of the posterior rami (L1-L5) was traced through LMT to determine its distribution. Using immunohistochemistry, Type I/II muscle fibers were identified in 29 muscle biopsies from one fresh frozen specimen. The total area and number of each cell type was calculated using Visiopharm® (image analysis software). Architectural and fiber type data were analyzed using ANOVA with Tukey’s post-hoc test (p ≤ 0.05). From L1-L4, LMT had three architecturally distinct regions: superficial, intermediate and deep. At L5, intermediate LMT was absent. Mean FBL decreased significantly from superficial (5.8 ± 1.6cm) to deep regions (2.9 ± 1.1cm) as did volume (superficial, 5.6 ± 2.3ml; deep, 0.7 ± 0.3ml). In contrast, mean FBA increased from superficial to deep. The medial branch of the posterior ramus (L1-L5) supplied the five bands of LMT. Each medial branch in turn divided to supply the deep, intermediate and superficial regions separately. The area occupied by Type I fibers was significantly less (p< 0.01) in the deep (56%) compared with the superficial regions (75%). Based on architecture and morphology, superficial LMT with the longest FBL and relatively small FBA is well designed for torque production and controlling the lumbar lordosis. Intermediate LMT with significantly longer FBL compared with the deep region and with its caudal to cranial line of action may help to control intersegmental stability. Furthermore, the absence of intermediate LMT at L5 and may contribute to the higher incidence of instability observed at the lumbosacral junction. Deep LMT with its short FBL, large FBA and proximity to the axis of spinal rotation may function to provide proprioceptive input to the CNS rather than a primary stabilizer of the lumbar spine.

The purpose of the study was to investigate the static and dynamic architecture of supraspinatus throughout its volume in the normal and pathological state. The architecture was first investigated in cadaveric specimens free of any tendon pathology. Using a serial dissection and digitization method tailored for supraspinatus, the musculotendinous architecture was modeled in situ. The 3D model reconstructed in Autodesk MayaTM allowed for visualization and quantification of the fiber bundle architecture i.e. fiber bundle length (FBL), pennation angle (PA), muscle volume (MV) and tendon dimensions. Based on attachment sites and architectural parameters, the supraspinatus was found to have two architecturally distinct regions, anterior and posterior, each with three subdivisions. The findings from the cadaveric investigation served as a map and platform for the development of an ultrasound (US) protocol that allowed for the dynamic fiber bundle architecture to be quantified in vivo in normal subjects and subjects with a full-thickness supraspinatus tendon tear. The architecture was studied in the relaxed state and in three contracted states (60º abduction with either neutral rotation, 80º external rotation, or 80º internal rotation). The dynamic changes in the architecture within the distinct regions of the muscle were not uniform and varied as a function of joint position. Mean FBL in the anterior region shortened significantly with contraction (p<0.05) but not in the posterior. In the anterior region, mean PA was significantly smaller in the middle part compared to the deep (p<0.05). Comparison of the normal and pathological muscle found large differences in the percentage change of FBL and PA with contraction. The architectural parameter that showed the largest changes with tendon pathology was PA. In sum, the results showed that the static and dynamic fiber bundle architecture of supraspinatus is heterogeneous throughout the muscle volume and may influence tendon stresses. The architectural data collected in this study and the 3D muscle model can be used to develop future contractile models. The US protocol may serve as an assessment tool to predict the functional outcome of rehabilitative exercises and surgery.

Une étude multi-échelle du gonflement d’une smectite en milieu confiné est proposée par une analyse de la pression de gonflement et de la porosité à différentes échelles (inter-agrégat, inter-particulaire, interfoliaire). Un dispositif spécifique, œdo-tomomètre, est conçu pour assurer le suivi couplé de la pression de gonflement et de la porosité inter-agrégat identifiable par microtomographie de rayon X (taille > 5 µm). D’autres analyses par adsorption de gaz et en microscopie permettent de caractériser la porosité à plus petites échelles. Différents types de fluides sont utilisés pour hydrater les éprouvettes afin de contrôler les composantes cristalline et osmotique du gonflement (solutions de NaCl à différentes forces ioniques et solution de méthacrylate de méthyle (MMA)). Les résultats obtenus en comparant notamment les éprouvettes hydratées par des solutions ioniques de NaCl à celles hydratées par du MMA montrent que l’évolution de la pression de gonflement et la réorganisation de la microstructure (diminution en taille de la porosité inter-agrégat, fractionnement possible des particules) sont intimement liées. Par ailleurs, ces résultats comparés à ceux calculés par des modèles phénoménologiques (BExM par exemple, modèle de Barcelone pour les sols gonflants) sont assez bien corrélés / A multi-scale investigation of the swelling behaviour of a purified bentonite under constrained conditions is carried out by a monitoring of the swelling pressure and the porosity evolution at various scales (inter-aggregate, inter-particle, interlayer). A specific oedometer, oedo-tomometer, has been designed to monitor both the swelling pressure and the inter-aggregate porosity evolution deduced from microfocus X-ray computed tomography measurements (size higher than 5 µm). Further investigations of the porosity at lower scale were conducted at various stages of the swelling process by gas adsorption mainly and transmission electron microscopy. Various fluids were used for specimen hydration to control either crystalline swelling or osmotic swelling (NaCl solutions at different ionic strengths and methyl methacrylate (MMA) solution). The results, which combined both swelling pressure measurements and quantification of microstructure evolution upon hydration for the two different solutions, give sound understanding on the development of osmotic and/or crystalline swelling and their relative impact both on the microstructure and on the magnitude of the macroscopic swelling pressure of compacted smectites. Furthermore, these results compared to calculated results (BExM for example) shows similar evolutions

Les décharges pulsées nanoseconde sont caractérisées par un fort champ réduit (centaine Td) et une forte thermodynamique hors-équilibre. Ils ont de l’énergie électronique de quelques eV à quelques dizaine eV et la déposition d’énergie spécifique de 10^-3 eV/mol à quelques eV/mol.Deux configurations particulières sont étudiées: (i) décharge capillaire nanoseconde (nCD) fonctionnant à la pression modérée et à une haute, et (ii) décharge contrôlée par barrière diélectrique surfacique de nanoseconde, fonctionnant à la pression atmosphérique ou à la pression plus élevée, et à une énergie spécifique de la déposition relativement faible.La décharge capillaire nanoseconde est un outil expérimental pour analyser le nanoseconde plasma dans certaines conditions extrêmes. Des expériences récentes de nCD ont révélé que la cinétique du plasma change considérablement quand l’énergie spécifique de la déposition est plus élevée. L'un des objectifs du travail est d'étudier numériquement les effets de la cinétique modifiée sur la technique classique de la mesure de l’actinométrie, et sur l’évolution spatial-temporelle du plasma dans la décharge et en post-décharge.La décharge contrôlée par barrière diélectrique surfacique de nanoseconde a été largement étudiée dans la communauté de l'aérodynamique. Cependant, au début de démarrer le travail, les paramètres de nSDBD n'étaient pas bien compris, la comparaison des calculs numériques et des résultats expérimentaux n'étaient pas disponibles. Par conséquent, la modélisation de nSDBD et la comparaison avec des résultats expérimentaux sous les mêmes paramètres est un autre objet de la thèse.Le travail dans ce manuscrit est organisé en trois parties. Dans la première partie, la modélisation numérique et les expérimentations de l’actinométrie basée sur l’Ar sont utilisées pour étudier la densité de l’oxygène atomique dans une décharge capillaire nanoseconde. Un schéma cinétique décrivant le comportement cohérent de l'ensemble des données expérimentales est développé. Les processus principaux, qui sont responsables de la population et de la décroissance des trois espèces intéressés sont sélectionnés à base de l’analyse de la sensibilité et du taux. Le rôle des réactions entre les espèces excitées et les électrons a l’entrée de la post-décharge pour une décharge pulsée au grand champ électrique et haute énergie spécifique de la déposition est discuté.La deuxième partie est consacrée à étudier, analyser et prévoir des caractéristiques de la décharge et de la post-décharge de nCD sous différents énergie spécifique de la déposition , basées sur un code auto - cohérent bidimensionnel, nonPDPsim. La propagation de la décharge a été modélisée. Deux modes de propagation ont été identifiés, trois formes d'ondes d'ionisation sont trouvées en variant le rayon de tube. Le taux de décroissance et la distribution radiale des électrons et de N2(C3) dans la post-décharge sont étudiés en respectant l’énergie spécifique de la déposition.Finalement, un modèle parallèle bidimensionnel PASSKEy («PArallel Streamer Solver with KinEtics») a été développé et validé pour modéliser le nSDBD. Une série de calculs numériques pour un seul pulse de nSDBD dans l'air à la pression atmosphérique à une amplitude de tension de 24 kV a été effectuée, les résultats ont été comparés avec des résultats expérimentaux dans les mêmes conditions. Vitesse calculée et mesurée De l’entrée de la décharge, courant électrique, carte plan 2D d'émission de N2(C3)->N2(B3), et perturbations hydrodynamiques provoquées par la décharge sur l'échelle de temps sont analysées de 0, 2 à 5us. L'effet de différents processus cinétiques dans la distribution 2D de la chaleur est étudié. Les données sont présentées et analysées pour la polarité de tension négative et positive. Un ensemble de calculs paramétriques avec différentes permittivité diélectrique, différent épaisseur des diélectriques et différente pression ambiantes sont présentés. / Nanosecond pulsed discharges are characterized by high reduced electric fields (hundreds of Td) and strong nonequilibrium. They have characteristic electron energies of a few to tens of eV and specific energy deposition ranging from 10$^{-3}$ eV/mol to a few eV/mol. The energetic electrons can efficiently generate chemical active species, lead to fast gas heating. These discharges are found in a growing list of successful practical applications: gas pollution control, surface treatment, plasma assisted aerodynamics, plasma assisted biology and medicine and plasma assisted combustion.Two particular configurations are studied in present work: (i) nanosecond capillary discharge (nCD) operated at moderate pressures and high specific deposited energy, and (ii) nanosecond surface dielectric barrier discharge (nSDBD) operated at atmospheric or higher pressures and relatively low specific deposited energy.Nanosecond capillary discharge is an experimental tool to analyse nanosecond plasma in some limit extreme conditions. Recent nCD experiments revealed that, plasma kinetics changes dramatically at high specific energy deposition. One of the aims of the present work, is to study numerically the effects of the changed kinetics to the classical actinometry measurement technique, and the spatial-temporal evolution of plasmas during discharge and afterglow.Nanosecond surface dielectric barrier discharge has been widely studied in the community of aerodynamics. However, at the moment of starting the thesis, the parameters of nSDBD plasma were not yet clearly understood, detailed comparison of numerical calculations and experiments were not available. Therefore, modelling of nSDBD and comparison with experiments performed for the same parameters is another object of the presented thesis.The results in the thesis are presented in three parts. In the first part, numerical modelling and experiment of Ar-based actinometry are used to study the atomic oxygen density in nanosecond capillary discharge. A kinetic scheme describing consistent behavior of the set of the experimental data is developed. The main processes responsible for population and decay of the three species of interest are selected on the basis of sensitivity and rate analysis. The role of the reactions between excited species and electrons in early afterglow for pulsed discharges at high electric fields and high values of specific deposited energy is discussed. Density of O-atoms in the ground state is obtained from the calculations.The second part is devoted to study, analyse and predict the features of the discharge and afterglow of nCD under different specific energy deposition based on a two--dimensional self-consistent code, nonPDPsim. Propagation of the discharge have been modelled. Two modes of propagation were identified, three shapes of ionization waves are found with various tube radius. The decay rate and radial distribution of electrons and $rm N_2(C^3Pi_u)$ in the afterglow are studied with respect to specific energy deposition.Finally, a two--dimensional parallel PASSKEy (``PArallel Streamer Solver with KinEtics'') code coupling plasma and hydrodynamics has been developed and validated to model nSDBD. Series of numerical calculations for a single pulse nSDBD in atmospheric pressure air at 24~kV voltage amplitude has been performed, the results were compared with experiments in the same conditions. Calculated and measured velocity of the discharge front, electrical current, 2D map of emission of N$_2$(C$^3Pi_u$) $rightarrow$ N$_2$(B$^3Pi_g$), and hydrodynamic perturbations caused by the discharge on the time scale $0.2-5$~$mu$s are analysed. The effect of different kinetics processes in 2D distribution of heat release is studied. The data are presented and analyzed for negative and positive polarity of voltages. A set of parametric calculations with different dielectric permittivity, the thickness of dielectric and ambient pressures are presented.