Found 1496 result(s)
Found 150 page(s)

Overview of security issues in Vehicular ad-hoc networks

  • Fuentes, José María de
  • González-Tablas, Ana Isabel
  • Ribagorda, Arturo
Vehicular ad-hoc networks (VANETs) are a promising communication scenario. Several new applications are envisioned, which will improve traffic management and safety. Nevertheless, those applications have stringent security requirements, as they affect road traffic safety. Moreover, VANETs face several security threats. As VANETs present some unique features (e.g. high mobility of nodes, geographic extension, etc.) traditional security mechanisms are not always suitable. Because of that, a plethora of research contributions have been presented so far. This chapter aims to describe and analyze the most representative VANET security developments.
Project:


Multiple openings of forward markets: experimental evidence

  • Ferreira, José Luis
  • Kujal, Praveen
  • Rassenti, Stephen
We test the strategic motive to sell forward in experimental Cournot duopoly and quadropoly environments with multiple forward markets. Using random matching, we test two versions of forward markets with finite (Allaz and Vila, 1993) and indefinite number of periods. We find that the results for the Allaz and Vila (1993) model are remarkably close to the predicted theoretical results for both duopolies and quadropolies. We then test a version of the model to allow for indefinitely many periods. There are multiple equilibria in this theoretical model, including both the competitive and collusive outcomes. We find that the initial "collusive hypothesis" is not ratified, and that outcomes are nearly competitive. Sales take place mostly in the first few openings of futures markets. Again, these results hold for both duopolies and quadropolies.
Project:


The effect of experience in Cournot play

  • Ferreira, José Luis
  • Kujal, Praveen
  • Rassenti, Stephen
Strategic play requires that players in oligopolies be more sophisticated than in perfectly competitive markets. It thus seems reasonable to assume that player experience becomes important as the environment gets more complicated. We find that subject experience indeed plays an important role. While inexperienced symmetric duopolies play around the Nash-Cournot quantity, experienced duopolies reduce output and get closer to the monopolistic outcome. Both inexperienced and experienced symmetric quadropolies,however, produce output above the Nash-Coumot equilibrium but, even in this case, output is lower for experienced quadropolies. Experience, however, does not make markets less competitive with the introduction of cost asymmetry. Under cost asymmetry, and relative to the equilibrium prediction, high cost firms produce more output than low cost firms. Analysis of individual data tells us that experienced duopolies and quadropolies adjust output in the same direction as their rivals. Due to the strategic substitutability of quantity choice, we interpret this as an attempt at tacitly colluding. This is true for both duopolies and quadropolies.
Project:


Reaction to public information in asset markets: does ambiguity matter?

  • Corgnet, Brice
  • Kujal, Praveen
  • Porter, David
We report experiments that examine trader reaction to ambiguity when dividend information is revealed sequentially. We find that experienced traders are better at internalizing ambiguity than inexperienced subjects. No significant differences are observed in the ambiguity versus control treatments regarding prices, price volatility and volumes for experienced subjects. However, relative to the control, prices are higher, volatility greater and trading unsophisticated for inexperienced subjects in the ambiguity treatment. Price changes are consistent with news revelation regardless of subject experience and the degree of ambiguity. Further, we do not find under or over price reactions to news. Regardless of experience, market reaction to news moves in line with fundamentals.
Project:


Interoperabilidad y repositorios: el Grupo de Trabajo SONEX, El grupo de trabajo SONEX para la notificación e intercambio de la producción académica: oportunidades para la incorporación automática de contenidos a los repositorios

  • Castro Martín, Pablo de
27 slides.-- Presentado en: X Workshop REBIUN sobre Proyectos Digitales (Valencia, Oct 7-8, 2010)., Durante el 'International Repositories Workshop' celebrado en Amsterdam en marzo de 2009 se señaló la urgente necesidad de incrementar los contenidos almacenados en los repositorios institucionales. Bien a través de la promoción de iniciativas para la transferencia automática de contenidos desde los editores o mediante el desarrollo de la interoperabilidad entre los sistemas de información sobre investigación o 'Current Research Information Systems' (CRIS) y los repositorios institucionales, era preciso fomentar estrategias que permitieran automatizar la incorporación de contenidos. El grupo de trabajo SONEX (Scholarly Output Notification and Exchange), financiado por el JISC (Joint Information Systems Committee) del Reino Unido, se estableció entonces para analizar las cuestiones relacionadas con la interoperabilidad de repositorios y contribuir a la implementación de las soluciones detectadas. Tras la publicación el pasado marzo de 2010 por parte del JISC de la convocatoria para financiar proyectos relacionados con el depósito de contenidos en los repositorios del Reino Unido, el papel del grupo SONEX se amplió para incluir la colaboración y el apoyo a los proyectos seleccionados en dicha convocatoria. En esta presentación se resume el marco de interoperabilidad de repositorios analizado por SONEX a lo largo del último año y se mencionan algunas iniciativas en la materia que vienen desarrollándose en diversos repositorios españoles., Joint Information Systems Committee (JISC)
Project:


Current marketing practices and market orientation in the context of an emerging economy : the case of Uruguay

  • Lado, Nora
  • Duque Zuluaga, Lola Cristina
  • Alvarez Bassi, Daniel
This research explores the link between contemporary marketing practices, market orientation and business performance in Uruguay, an emergent country that has recovered from an economic crisis. These approaches seem to be related, but there is no existing evidence to confirm this impression. Lessons can be learned from understanding how effective is the adoption of marketing practices under a crisis scenario. Using data from interviews with 143 micro and small enterprises’ managers, we identify three clusters dependant on the combination of marketing practices: a multi-marketing cluster, a medium-level relationship marketing cluster and a transactional cluster. A model relating market orientation components and various performance measures is tested for the three clusters, showing that the multi-marketing and transactional clusters are more effective in translating efforts and resources into business outcomes.
Project:


Similaridad y contraste mediante profundidad estadística

  • López López, Ángel
El objetivo principal de esta tesis doctoral es la exploración de las posibilidades que ofrecen las funciones de profundidad estadística, tanto en la definición de medidas de proximidad en sentido estadístico, como en problemas de clasificación y de contraste de bondad de ajuste., Financiación recibida de: Ministerio de Ciencia e Innovación, Ministerio de Educación y Comunidad de Madrid a través de los proyectos de investigación ECO2008-05080,SEJ2005-06454, BEC2002-03769 y 06/HSE/0181/2004.
Project:


Dynamic optimization techniques to enhance scalability and performance of MPI-based applications, Técnicas de optimización dinámicas de aplicaciones paralelas basadas en MPI

  • Filgueira Vicente, Rosa
Parallel computation on cluster architectures has become the most common solution for developing high-performance scientific applications. Message Passing Interface (MPI) [Mes94] is the message-passing library most widely used to provide communications in clusters. MPI provides a standard interface for operations such as point-to-point communication, collective communication, synchronization, and I/O operations. Along the I/O phase, the processes frequently access a common data set by issuing a large number of small non-contiguous I/O requests [NKP+96a, SR98], which might create bottlenecks in the I/O subsystem. These bottlenecks are still higher in commodity clusters, where commercial networks are usually installed. Many of those networks, such as Fast Ethernet or Gigabit, have high latency and low bandwidth which introduce performance penalties during the program execution. Scalability is also an important issue in cluster systems when many processors are used, which may cause network saturation and still higher latencies. As communication-intensive parallel applications spend a significant amount of their total execution time exchanging data between processes, the former problems may lead to poor performance not only in the I/O subsystem, but also in communication phase. Therefore, we can conclude that it is necessary to develop techniques for improving the performance of both communication and I/O subsystems. The main goal of this Ph.D. thesis is to improve the scalability and performance of MPI-based applications executed in clusters reducing the overhead of I/O and communications subsystems. In summary, this work proposes two techniques that solve these problems in an efficient way managing the high complexity of a heterogeneous environment: • Reduction in the number of communications in collective I/O operations: This thesis targets the reduction of the bottleneck in the I/O subsystem. Many applications use collective I/O operations to read/write data from/to disk. One of the most used is the Two-Phase I/O technique extended by Thakur and Choudhary in ROMIO. In this technique, many communications among the processes are performed, which could create a bottleneck. This bottleneck is still higher in commodity clusters, where commercial networks are usually installed, and in CMP clusters where the I/O bus is shared by the cores of a single node. Therefore, we propose improving locality in order to reduce the number of communications performed in Two-Phase I/O. • Reduction of transferred data volume: This thesis attemps to reduce the cost of interchanged messages, reducing the data volume by using lossless compression among processes. Furthermore, we propose turning compression on and off and selecting at run-time the most appropriate compression algorithms depending on the characteristics of each message, network performance, and compression algorithms behavior.-------------------------------------------------------------------------------------------------------------------------------------------------, En la actualidad, las aplicaciones utilizadas en los entornos de computación de altas prestaciones, como por ejemplo simulaciones científicas o aplicaciones dedicadas a la extracción de datos (data-mining), necesitan además de enormes recursos de cómputo y memoria, el manejo de ingentes volúmenes de información. Las arquitecturas cluster se han convertido en la solución más común para ejecutar este tipo de aplicaciones. La librería MPI (Message Passing Interface) [Mes94] es la más utilizada en estos entornos, ya que ofrece un interfaz estándar para operaciones de comunicación punto a punto, colectivas, sincronización y de E/S. Durante la fase de E/S de las aplicaciones, los procesos acceden a un gran conjunto de datos mediante pequeñas peticiones de datos no-contiguos, por lo que pueden provocar cuellos de botella en el sistema de E/S. Estos cuellos de botella, pueden ser todavía mayor en los cluster, ya que se suelen utilizar redes comerciales como Fast Ethernet o Gigabit, las cuales tienen una gran latencia y bajo ancho de banda. Por otra parte la escalabilidad es un importante problema en los clusters, cuando se ejecutan a la vez un gran número de procesos, ya que pueden causar saturación de la red, y aumenar la latencia. Como consecuencia de una comunicación intensiva, las aplicaciones gastan mucho tiempo intercambiando información entre los procesos, provocando problemas tanto en el sistema de comunicación, como en el de E/S. Por lo tanto, podemos concluir que en un cluster los subsistemas de E/S y de comunicaciones representan uno de los principales elementos en los que conviene mejorar su rendimiento. El principal objetivo de esta Tesis Doctoral es mejorar la escalabilidad y rendimientos de las aplicaciones MPI ejecutadas en arquitecturas cluster, reduciendo la sobrecarga de los sistemas de comunicación y de E/S. Como resumen, este trabajo propone dos técnicas para resolver estos problemas de forma eficiente: • Reducción del número de comunicaciones en la operaciones colectivas de E/S: Esta tesis tiene como uno de sus objetivos reducir los cuellos de botella producidos en el sistema de E/S. Muchas aplicaciones científicas utilizan operaciones colectivas de E/S para leer/escribir datos desde/al disco. Una de las técnicas más utilizas es Two-Phase I/O ampliada por Thakur and Choudhary en ROMIO. En esta técnica se realizan muchas comunicaciones entre los procesos, por lo que pueden crear un cuello de botella. Este cuello de botella es aún mayor en los cluster que tiene instaladas redes comerciales, y en los clusters multicore donde el bus de E/S es compartido por todos los cores de un mismo nodo. Por lo tanto, nosotros proponemos aumentar la localidad y disminuir a la vez en número de comunicaciones que se producen en Two-Phase I/O para reducir los problemas de E/S en las arquitecturas cluster. • Reducción del volumen de datos en las comunicaciones: Esta tesis propone reducir el coste de las comunicaciones utilizando técnicas de compresión sin perdida. Concretamente, proponemos activar y desactivar la compresión y elegir el algoritmo de compresión en tiempo de ejecución, dependiendo de las características de cada mensaje, de la red y del comportamiento de los algoritmos de compresión.
Project:


Evolutionary model of existing competition and voluntary disclosure

  • Núñez-Nickel, Manuel
  • Gago Rodríguez, Susana
We analyze how, in the absence of capital market incentives, the influence of existing competition on voluntary disclosure is an evolving process which has a non-monotonic design. The progressive capability of rivals to forecast significant information and the increasing losses of abnormal profits during the industry life cycle generate fears and incentives that change the sign of the relationship between competition and the probability of voluntary disclosure throughout the industry’s development. We support this new design empirically by applying a semi-parametric Cox model to 28 years of archival data for the entire Spanish newspaper sector. We also find that the best fitting model is the first harmonic of a Fourier series
Project:


The Hydrogen Laminar Jet

  • Sánchez-Sanz, Mario
  • Rosales, Marco
  • Sánchez, Antonio L.
Numerical and asymptotic methods are used to investigate the structure of the hydrogen jet discharging into a quiescent air atmosphere. The analysis accounts in particular for the variation of the density and transport properties with composition. The Reynolds number of the flow "Rj", based on the initial jet radius "a", the density and viscosity of the jet and the characteristic jet velocity, is assumed to take moderately large values, so that the jet remains slender and stable, and can be correspondingly described by numerical integration of the continuity, momentum and species conservation equations written in the boundary-layer approximation. The solution for the velocity and composition in the jet-development region of planar and round jets, corresponding to streamwise distances of order "Rj a", is computed numerically, along with the solutions that emerge both in the near field and in the far field. The small value of the hydrogen-to-airmolecular weight ratio is used to simplify the solution by considering the asymptotic limit of vanishing jet density. The development provides at leading order explicit analytical expressions for the far-field velocity and hydrogen mass fraction that describe accurately the hydrogen jet near the axis. The information provided can be useful in particular to characterize hydrogen discharge processes from holes and cracks., This work was supported by the SpanishMICINN under Project # ENE2008-06515-C04 and by the Comunidad de Madrid under Project S2009/ENE-1597 (HYSYCOMB), Publicado
Project:


Advanced search