Dada la creciente demanda de servicios multimedia en movilidad, las operadoras de telefonía móvil necesitan aumentar constantemente la capacidad de sus sistemas. En ocasiones la única solución posible para mejorar las prestaciones del sistema es incrementar el número de transmisores, siendo ésta una solución costosa. Además, en UMTS utilizar más emplazamientos macro-celulares no siempre garantiza una mayor capacidad del sistema, por los grandes problemas de interferencia mutua que sufren dos macro-celdas si se encuentran demasiado próximas. La única solución que se puede adoptar en estos casos es hacer uso de redes jerárquicas celulares, añadiendo micro-celdas que utilicen una frecuencia de portadora distinta. Esta Tesis Doctoral aporta una visión complementaria al despliegue de las redes jerárquicas y propone aprovechar dicho despliegue para modificar las redes de acceso radio UMTS con nueva arquitectura de nodos B múltiples o Multinodo B. El sistema a desarrollar utiliza la tecnología de nodos B UMTS a los que se conectan un número de antenas distribuidas en diferentes puntos del área de servicio, conformando el equivalente a un array distribuido espacialmente. La incorporación de técnicas de procesado en array junto con mecanismos de gestión dinámica de recursos y la tecnología de enlaces ópticos proporcionarán a este sistema una estructura de acceso radio óptima en entornos micro-celulares y en interiores, con gran flexibilidad y adaptabilidad. Las ventajas de utilizar esta configuración, que se detallan a lo largo de la Tesis Doctoral, son fundamentalmente la mejora de la capacidad de la red de acceso, minimización del impacto radioeléctrico y óptima gestión de recursos. En primer lugar, en esta Tesis Doctoral se demuestra que la nueva arquitectura radio es capaz de dotar al sistema UMTS de una diversidad espacial adicional, diversidad que mejorará los requerimientos de potencia recibida en los terminales, sobre todo en los escenarios micro-celulares. Respecto / Monserrat Del Río, JF. (2007). Mejora de la Arquitectura de Acceso Radio UMTS mediante Multinodos B [Tesis doctoral no publicada]. Universitat Politècnica de València. https://doi.org/10.4995/Thesis/10251/1937 / Palancia

Over the last few years, public transportation has become more desirable as capacity of existing roadways failed to keep up with rapidly increasing traffic demand. Buses are one of the most common modes of public transportation with low impact on network capacity, especially in small and congested urban areas. However, the use of regularly scheduled buses as the main public transport mode can become useless with the presence of traffic congestion and dense construction areas. In cases like these, innovative solutions, such as bus rapid transit (BRT), can provide an increased level of service without having to resort to other, more expensive modes, such as light rail transit (LRT) and metro systems (subways). Transit signal priority (TSP), which provides priority to approaching buses at signalized intersections by extending the green or truncating the red, can also increase the performance of the bus service. Understanding the combined impact of TSP and BRT on network traffic operations can be complex. Although TSP has been implemented worldwide, none of the previous studies have examined in depth the effects of using conditional and unconditional TSP strategies with a BRT system. The objective of this research is to evaluate the effectiveness of BRT without TSP, then with conditional or unconditional TSP strategies. The micro-simulation software VISSIM was used to compare different TSP and BRT scenarios. These simulation scenarios include the base scenario (before implementation of the TSP and BRT systems), Unconditional TSP (TSP activates for all buses), Conditional TSP 3 minutes behind (TSP only activates for buses that are 3 minutes or more behind schedule), Conditional TSP 5 minutes behind (only activates for buses 5 minutes or more behind schedule), BRT with no TSP, BRT with Unconditional TSP, BRT with Conditional TSP 3 minutes behind, and BRT with Conditional TSP 5 minutes behind. The VISSIM simulation model was developed, calibrated and validated using a variety of data that was collected in the field. These data included geometric data, (number of lanes, intersection geometries, etc.); traffic data (average daily traffic volumes at major intersections, turning movement percentages at intersections, heavy vehicle percentages, bus passenger data, etc.); and traffic control data (signal types, timings and phasings, split history, etc.). Using this field data ensured the simulation model was sufficient for modeling the test corridor. From this model, the main performance parameters (for all vehicles and for buses only) for through movements in both directions (eastbound and westbound) along the corridor were analyzed for the various BRT/TSP scenarios. These parameters included average travel times, average speed profiles, average delays, and average number of stops. As part of a holistic approach, the effects of BRT and TSP on crossing street delay were also evaluated. Simulation results showed that TSP and BRT scenarios were effective in reducing travel times (up to 26 %) and delays (up to 64%), as well as increasing the speed (up to 47%), compared to the base scenario. The most effective scenarios were achieved by combining BRT and TSP. Results also showed that BRT with Conditional TSP 3 minutes behind significantly improved travel times (17 – 26%), average speed (30 – 39%), and average total delay per vehicle (11 – 32%) for the main corridor through movements compared with the base scenario, with only minor effects on crossing street delays. BRT with Unconditional TSP resulted in significant crossing street delays, especially at major intersections with high traffic demand, which indicates that this scenario is impractical for implementation in the corridor. Additionally, BRT with Conditional TSP 3 minutes behind had better travel time savings than BRT with Conditional TSP 5 minutes behind for both travel directions, making this the most beneficial scenario. This research provided an innovative approach by using nested sets (hierarchical design) of TSP and BRT combination scenarios. Coupled with microscopic simulation, nested sets in the hierarchical design are used to evaluate the effectiveness of BRT without TSP, then with conditional or unconditional TSP strategies. The robust methodology developed in this research can be applied to any corridor to understand the combined TSP and BRT effects on traffic performance. Presenting the results in an organized fashion like this can be helpful in decision making. This research investigated the effects of BRT along I-Drive corridor (before and after conditions) at the intersection level. Intersection analysis demonstrated based on real life data for the before and after the construction of BRT using the Highway Capacity SoftwareTM (HCS2010) that was built based on the Highway Capacity Manual (HCM 2010) procedures for urban streets and signalized intersections. The performance measure used in this analysis is the level of service (LOS) criteria which depends on the control delay (seconds per vehicle) for each approach and for the entire intersection. The results show that implementing BRT did not change the LOS. However, the control delay has improved at most of the intersections' approaches. The majority of intersections operated with an overall LOS "C" or better except for Kirkman Road intersection (T2) with LOS "E" because it has the highest traffic volumes before and after BRT construction. This research also used regression analysis to observe the effect of the tested scenarios analyzed in VISSIM software compared to the No TSP – No BRT base model for all vehicles and for buses only. The developed regression model can predict the effect of each scenario on each studied Measures of Performance (MOE). Minitab statistical software was used to conduct this multiple regression analysis. The developed models with real life data input are able to predict how proposed enhancements change the studied MOEs. The BRT models presented in this research can be used for further sensitivity analysis on a larger regional network in the upcoming regional expansion of the transit system in Central Florida. Since this research demonstrated the operational functionality and effectiveness of BRT and TSP systems in this critical corridor in Central Florida, these systems' accomplishments can be expanded throughout the state of Florida to provide greater benefits to transit passengers. Furthermore, to demonstrate the methodology developed in this research, it is applied to a test corridor along International Drive (I-Drive) in Orlando, Florida. This corridor is key for regional economic prosperity of Central Florida and the novel approach developed in this dissertation can be expanded to other transit systems.

Les principales caractéristiques neuropathologiques de la maladie d’Alzheimer (MA) sont les plaques séniles extracellulaires composées de peptide amyloïde β (Aβ) et les enchevêtrements neurofibrillaires intracellulaires composés de Tau hyperphosphorylé. Les mécanismes conduisant à la formation de ces lésions sont encore peu connus et le laboratoire a récemment caractériser le gène “bridging integrator 1” (BIN1), deuxième facteur de risque génétique le plus associé au risque de MA, comme facteur de risque potentiellement associé à la pathologie Tau. Une interaction entre les deux protéines a été décrite in vitro et in vivo suggérant que BIN1 pourrait être impliqué dans le développement de la pathologie associée à Tau dans le cadre de la MA. Cependant, ce rôle de l'interaction BIN1-Tau dans le processus pathophysiologique de la MA n'est pas connu et il reste ainsi à déterminer si cette interaction constitue une cible thérapeutique potentielle. Ce projet a visé alors à mieux comprendre les acteurs de cette interaction en identifiant les modulateurs et les voies moléculaires impliquées dans le contrôle de l'interaction BIN1-Tau, puis de déterminer comment cette interaction est modulée dans le contexte de la MA. Nous avons utilisé pour cela des approches complémentaires de biochimie, de résonance magnétique nucléaire et de microscopie confocale. Comme modèle cellulaire, des cultures primaires de neurones de rat ont été utilisées, et la méthode “proximity ligation assay” (PLA) a été développée comme approche principale pour observer l'interaction BIN1-Tau dans ces cellules. Nous avons déterminé que l'interaction se produit entre les domaines SH3 de BIN1 et le PRD de Tau et nous avons démontré que l’interaction est modulée par la phosphorylation de Tau et BIN1: la phosphorylation de la Thréonine 231 de Tau diminue son interaction avec BIN1, tandis que la phosphorylation de BIN1 à la Thréonine 348 (T348) augmente son interaction avec Tau. Nous avons mis au point une approche de criblage d’haut contenu semi-automatisée et basé sur une bibliothèque de composés commerciaux. Ce criblage s’est basé sur des cultures primaires de neurones comme modèle cellulaire et le PLA pour détecter l'interaction BIN1-Tau. Nous avons identifié plusieurs composés capables de moduler l'interaction BIN1-Tau, notamment U0126, un inhibiteur de MEK-1/2, qui diminue cette interaction, et la cyclosporine A, un inhibiteur de la calcineurine, qui au contraire augmente celle-ci en augmentant la phosphorylation de T348 de BIN1. Par ailleurs les “Cyclin-dependent kinases” (CDK) ont été montré comme contrôlant aussi ce site de phosphorylation. Nous avons donc mis en évidence le couple Calcineurine/CDK comme contrôlant la phosphorylation T348 de Bin1 et donc l’interaction BIN1-Tau. Nous avons également développé un modèle murin de tauopathie dans lequel nous avons surexprimé BIN1 humain. Nous avons observé que la surexpression de BIN1 résorbait les déficits de mémoire à long terme et réduisait la présence d'inclusions intracellulaires de Tau phosphorylée, provoquées par la surexpression de Tau, ce qui était associé à une augmentation de l'interaction BIN1-Tau. En utilisant des échantillons de cerveau humain post-mortem, nous avons observé que les niveaux de l’isoforme BIN1 neuronal étaient diminués dans les cerveaux d’AD, alors que les niveaux relatifs de BIN1 phosphorylé à T348 étaient augmentés, suggérant un mécanisme compensatoire. Cette étude a démontré la complexité et la dynamique de l’interaction BIN1-Tau dans les neurones, a révélé des modulateurs et des voies moléculaires potentiellement impliquées dans cette interaction, et a montré que les variations de l’expression ou de l’activité de BIN1 ont des effets directs sur l’apprentissage et la mémoire, possiblement liés à la régulation de son interaction avec Tau. / The main neuropathological hallmarks of Alzheimer’s disease (AD) are the extracellular senile plaques composed of amyloid-β peptide (Aβ) and the intracellular neurofibrillary tangles composed of hyperphosphorylated Tau. The mechanisms leading to the formation of these lesions is not well understood and our lab has recently characterized the bridging integrator 1 (BIN1) gene, the second most associated genetic risk factor of AD and the first genetic risk factor to have a potential link to Tau pathology. The interaction between BIN1 and Tau proteins has been described in vitro and in vivo, which suggests that BIN1 might help us to understand Tau pathology in the context of AD. However, the role of BIN1-Tau interaction in the pathophysiological process of AD is not known, and whether this interaction is a potential therapeutic target remains to be determined. The aim of this project is to better understand the actors of BIN1-Tau interaction through the identification of the modulators and the molecular pathways involved therein, as well as to understand how BIN1-Tau interaction is modulated in the context of AD. We employed biochemistry, nuclear magnetic resonance, and confocal microscopy. We used rat primary neuronal cultures (PNC) as the cellular model and developed the proximity ligation assay (PLA) as the main readout of the BIN1-Tau interaction in cultured neurons. We determined that the interaction occurs between BIN1’s SH3 domain and Tau’s PRD domain, and demonstrated that it is modulated by Tau and BIN1 phosphorylation: phosphorylation of Tau at Threonine 231 decreases its interaction with BIN1, while phosphorylation of BIN1 at Threonine 348 (T348) increases its interaction with Tau. We developed a novel, semi-automated high content screening (HCS) assay based on a commercial compound library, also using PNC as the cellular model and PLA as the readout of BIN1-Tau interaction. We identified several compounds that are able to modulate the BIN1-Tau interaction, most notably U0126, an inhibitor of MEK-1/2, which reduced the interaction, and Cyclosporin A, an inhibitor of Calcineurin, which increased the interaction through increasing the BIN1 phosphorylation at T348. Furthermore, Cyclin-dependent kinases (CDK) were also shown as regulator of this phosphorylation site. These results suggest that the couple Calcineurin/CDK regulates BIN1 phosphorylation at T348 and consequently the BIN1-Tau interaction. We also developed a mouse model of tauopathy in which we overexpressed human BIN1. We observed that the overexpression of BIN1 rescued the long-term memory deficits and reduced the presence of intracellular inclusions of phosphorylated Tau, caused by Tau overexpression, and this was associated with an increase of BIN1-Tau interaction. Also, using post-mortem human brain samples, we observed that the levels of the neuronal BIN1 isoform were decreased in AD brains, whereas the relative levels of BIN1 phosphorylated at T348 were increased, suggesting a compensatory mechanism. Altogether, this study demonstrated the complexity and the dynamics of BIN1-Tau interaction in neurons, revealed modulators of and molecular pathways potentially involved in this interaction, and showed that variations in BIN1 expression or activity have direct effects on learning and memory, possibly linked to the regulation of its interaction with Tau.