High Performance FPGA-oriented mersenne twister uniform random number generator

Mersenne Twister (MT) uniform random number generators are key cores for hardware acceleration of Monte Carlo simulations. In this work, two different architectures are studied: besides the classical table-based architecture, a different architecture based on a circular buffer and especially targeting FPGAs is proposed. A 30% performance improvement has been obtained when compared to the fastest previous work. The applicability of the proposed MT architectures has been proven in a high performance Gaussian RNG.

Pseudo-random single photon counting for space-borne atmospheric sensing applications

The ability to accurately observe the Earth's carbon cycles from space gives scientists an important tool to analyze climate change. Current space-borne Integrated-Path Differential Absorption (IPDA) Iidar concepts have the potential to meet this need. They are mainly based on the pulsed time-offlight principle, in which two high energy pulses of different wavelengths interrogate the atmosphere for its transmission properties and are backscattered by the ground. In this paper, feasibility study results of a Pseudo-Random Single Photon Counting (PRSPC) IPDA lidar are reported. The proposed approach replaces the high energy pulsed source (e.g. a solidstate laser), with a semiconductor laser in CW operation with a similar average power of a few Watts, benefiting from better efficiency and reliability. The auto-correlation property of Pseudo-Random Binary Sequence (PRBS) and temporal shifting of the codes can be utilized to transmit both wavelengths simultaneously, avoiding the beam misalignment problem experienced by pulsed techniques. The envelope signal to noise ratio has been analyzed, and various system parameters have been selected. By restricting the telescopes field-of-view, the dominant noise source of ambient light can be suppressed, and in addition with a low noise single photon counting detector, a retrieval precision of 1.5 ppm over 50 km along-track averaging could be attained. We also describe preliminary experimental results involving a negative feedback Indium Gallium Arsenide (InGaAs) single photon avalanche photodiode and a low power Distributed Feedback laser diode modulated with PRBS driven acoustic optical modulator. The results demonstrate that higher detector saturation count rates will be needed for use in future spacebourne missions but measurement linearity and precision should meet the stringent requirements set out by future Earthobserving missions.

Linear Bayes policy for learning in contextual-bandits

Machine and Statistical Learning techniques are used in almost all online advertisement systems. The problem of discovering which content is more demanded (e.g. receive more clicks) can be modeled as a multi-armed bandit problem. Contextual bandits (i.e., bandits with covariates, side information or associative reinforcement learning) associate, to each specific content, several features that define the “context” in which it appears (e.g. user, web page, time, region). This problem can be studied in the stochastic/statistical setting by means of the conditional probability paradigm using the Bayes’ theorem. However, for very large contextual information and/or real-time constraints, the exact calculation of the Bayes’ rule is computationally infeasible. In this article, we present a method that is able to handle large contextual information for learning in contextual-bandits problems. This method was tested in the Challenge on Yahoo! dataset at ICML2012’s Workshop “new Challenges for Exploration & Exploitation 3”, obtaining the second place. Its basic exploration policy is deterministic in the sense that for the same input data (as a time-series) the same results are obtained. We address the deterministic exploration vs. exploitation issue, explaining the way in which the proposed method deterministically finds an effective dynamic trade-off based solely in the input-data, in contrast to other methods that use a random number generator.

Implementación de instrumentación virtual con LaBVIEW

El objetivo del PFC es el diseño e implementación de una aplicación que funcione como osciloscopio, analizador de espectro y generador de funciones virtual, todo dentro de la misma aplicacion. Mediante una tarjeta de adquisición de datos tomaremos muestras de señales del mundo real (sistema analógico) para generar datos que puedan ser manipulados por un ordenador (sistema digital). Con esta misma tarjeta también se podrán generar señales básicas, tales como señales senoidales, cuadradas.... y además se ha añadido la funcionalidad de generar señales moduladas en frecuencia, señales tipo Chirp (usadas comúnmente tanto en aplicaciones sonar y radar, como en transmisión óptica) o PRN (ruido pseudo-aleatorio que consta de una secuencia determinista de pulsos que se repite cada periodo, usada comúnmente en receptores GPS), como también señales ampliamente conocidas como el ruido blanco Gaussiano o el ruido blanco uniforme. La aplicación mostrará con detalle las señales adquiridas y analizará de diversas maneras esas señales. Posee la función de enventanado de los tipos de ventana mas comunes, respuesta en frecuencia, transformada de Fourier, etc. La configuración es elegida por el usuario en un entorno amigable y de visualización atractiva. The objective of the PFC is the design and implementation of an application that works as oscilloscope, spectrum analyzer and virtual signal generator, all within the same application. Through a data acquisition card, the user can take samples of real-world signals (analog system) to generate data that can be manipulated by a computer (digital system). This same card can also generate basic signals, such as sine waves, square waves, sawtooth waves.... and further has added other functionalities as frequency modulated signals generation, Chirp signals type generation (commonly used in both sonar and radar applications, such as optical transmission) or PRN (pseudo-random noise sequence comprising a deterministic pulse that repeats every period, commonly used in GPS receivers). It also can generate widely known as Gaussian white noise signals or white noise uniform signals. The application will show in detail the acquired signals and will analyze these signals in different ways selected by the user. Windowing function has the most common window types, frequency response, Fourier transform are examples of what kind of analyzing that can be processed. The configuration is chosen by the user throught friendly and attractive displays and panels.

A fitting algorithm for random modeling the PLC channel

The characteristics of the power-line communication (PLC) channel are difficult to model due to the heterogeneity of the networks and the lack of common wiring practices. To obtain the full variability of the PLC channel, random channel generators are of great importance for the design and testing of communication algorithms. In this respect, we propose a random channel generator that is based on the top-down approach. Basically, we describe the multipath propagation and the coupling effects with an analytical model. We introduce the variability into a restricted set of parameters and, finally, we fit the model to a set of measured channels. The proposed model enables a closed-form description of both the mean path-loss profile and the statistical correlation function of the channel frequency response. As an example of application, we apply the procedure to a set of in-home measured channels in the band 2-100 MHz whose statistics are available in the literature. The measured channels are divided into nine classes according to their channel capacity. We provide the parameters for the random generation of channels for all nine classes, and we show that the results are consistent with the experimental ones. Finally, we merge the classes to capture the entire heterogeneity of in-home PLC channels. In detail, we introduce the class occurrence probability, and we present a random channel generator that targets the ensemble of all nine classes. The statistics of the composite set of channels are also studied, and they are compared to the results of experimental measurement campaigns in the literature.

Photovoltaic LiNbO3particles: Applications to Biomedicine/Biophotonics

Recently, a novel method to trap and pattern ensembles of nanoparticles has been proposed and tested. It relies on the photovoltaic (PV) properties of certain ferroelectric crystals such as LiNbO3 [1,2]. These crystals, when suitably doped, develop very high electric fields in response to illumination with light of suitable wavelength. The PV effect lies in the asymmetrical excitation of electrons giving rise to PV currents and associated space-charge fields (photorefractive effect). The field generated in the bulk of the sample propagates to the surrounding medium as evanescent fields. When dielectric or metal nanoparticles are deposited on the surface of the sample the evanescent fields give rise to either electrophoretic or dielectrophoretic forces, depending on the charge state of the particles, that induce the trapping and patterning effects [3,4]. The purpose of this work has been to explore the effects of such PV fields in the biology and biomedical areas. A first work was able to show the necrotic effects induced by such fields on He-La tumour cells grown on the surface of an illuminated iron-doped LiNbO3 crystal [5]. In principle, it is conceived that LiNbO3 nanoparticles may be advantageously used for such biomedical purposes considering the possibility of such nanoparticles being incorporated into the cells. Previous experiments using microparticles have been performed [5] with similar results to those achieved with the substrate. Therefore, the purpose of this work has been to fabricate and characterize the LiNbO3 nanoparticles and assess their necrotic effects when they are incorporated on a culture of tumour cells. Two different preparation methods have been used: 1) mechanical grinding from crystals, and 2) bottom-up sol-gel chemical synthesis from metal-ethoxide precursors. This later method leads to a more uniform size distribution of smaller particles (down to around 50 nm). Fig. 1(a) and 1(b) shows SEM images of the nanoparticles obtained with both method. An ad hoc software taking into account the physical properties of the crystal, particullarly donor and aceptor concentrations has been developped in order to estimate the electric field generated in noparticles. In a first stage simulations of the electric current of nanoparticles, in a conductive media, due to the PV effect have been carried out by MonteCarlo simulations using the Kutharev 1-centre transport model equations [6] . Special attention has been paid to the dependence on particle size and [Fe2+]/[Fe3+]. First results on cubic particles shows large dispersion for small sizes due to the random number of donors and its effective concentration (Fig 2). The necrotic (toxicity) effect of nanoparticles incorporated into a tumour cell culture subjected to 30 min. illumination with a blue LED is shown in Fig.3. For each type of nanoparticle the percent of cell survival in dark and illumination conditions has been plot as a function of the particle dilution factor. Fig. 1a corresponds to mechanical grinding particles whereas 1b and 1c refer to chemically synthesized particles with two oxidation states. The light effect is larger with mechanical grinding nanoparticles, but dark toxicity is also higher. For chemically synthesized nanoparticles dark toxicity is low but only in oxidized samples, where the PV effect is known to be larger, the light effect is appreciable. These preliminary results demonstrate that Fe:LiNbO· nanoparticles have a biological damaging effect on cells, although there are many points that should be clarified and much space for PV nanoparticles optimization. In particular, it appears necessary to determine the fraction of nanoparticles that become incorporated into the cells and the possible existence of threshold size effects. This work has been supported by MINECO under grant MAT2011-28379-C03.

Contexto, criptoanálisis y propuesta de solución de la inscripción de la talla (original) de la Virgen de Candelaria de Tenerife (Canarias, España)

Esta investigación es un ejemplo de simbiosis entre criptoanálisis y desciframiento de lenguas. Es la búsqueda del sentido de una inscripción, un conjunto de casi doscientas letras latinas, en una talla de la Virgen María que estaba en la isla de Tenerife, en la localidad hoy de Candelaria, en las islas Canarias. La imagen desapareció en un temporal en el año 1826. No obstante, es posible lograr una gran certeza sobre qué letras tenía, acudiendo a las fuentes documentales textuales y artísticas. El conocimiento del significado, si lo hubiera, de la inscripción mariana, creemos que no puede lograrse sin la adecuada comprensión del contexto. Esto significa indagar en la historia de la misma talla, que se remonta hasta el siglo XIV o XV, en el estudio de la población autóctona canaria, así como de los pueblos que allí llegaron en sus diferentes momentos históricos. Además, es necesario conocer el redescubrimiento del archipiélago canario y sus procesos de conquista y evangelización. Todos estos datos irán ofreciendo un panorama nuevo y sorprendente para comprender no sólo las letras sino la misma imagen escultórica en madera. A partir de este momento la indagación se moverá en ver si las letras corresponden a alguna lengua posible, lo que nos ha llevado a analizar un amplísimo conjunto de textos lo más cercanos a la época bajo estudio, pertenecientes a alrededor de un centenar de lenguas. Tras el examen lingüístico se ha procedido a un estudio de las posibles formas criptográficas que se hubieran utilizado para generar el texto de la inscripción. Se ofrece un detallado y minucioso elenco de técnicas posibles que pudieran haberse adoptado y se criptoanaliza con exhaustividad las letras de la talla mariana. Al mismo tiempo se ofrece un nuevo marco criptológico de métodos y sistemas más ordenado y completo que el que hasta ahora venía considerándose, en especial desde el surgimiento de la criptografía de clave asimétrica. Continuamos la investigación sopesando la posible generación pseudo-aleatoria del texto, un texto que pudiera no tener sentido alguno. En este momento, y habiendo completado todas las posibilidades e hipótesis, habiéndose negado todas, volvemos a reconsiderar el cuerpo de conjeturas y supuestos. Desde ahí analizamos en profundidad el ámbito etnográfico y lingüístico bereber como hipótesis más plausible y probable. Tras la profundización en esta lengua y la corrección de los errores que nos llevaron a no detectarla en nuestro análisis precedente, llegamos a la conclusión de encontrarnos ante una lengua arcaica bereber, un conjunto de letras pertenecientes a una lengua y familia hoy no desaparecida, si bien muy modelada y difuminada por otras lenguas, en especial el árabe. Esto nos llevará a rescatar aspectos léxicos, morfológicos, sintácticos y fonéticos de este habla arcaica. Con todos estos datos realizamos un amplio estudio semántico de la talla tanto desde la perspectiva aborigen autóctona como cristiana. Finalmente, desde las voces lexicales y sus raíces de las lenguas bereberes e insulares amazigh, ofrecemos el significado de las letras inscritas en la talla mariana de Candelaria. ABSTRACT This research is an example of symbiosis between cryptanalysis and deciphering of languages. It is the search for meaning in an inscription, a group of about two hundred latin letters on a carving of the Virgin Mary that was on the island of Tenerife, in the town of Candelaria today, in the Canary islands. The image disappeared in a storm in 1826. However, it is possible to achieve a great certainty about what letters had, going to the textual and artistic documentary sources. The knowledge of the meaning, if any, of the marian inscription, can not be achieved without an adequate knowledge of the context. This means researching into the history of the same carving, which dates back to the fourteenth and fifteen century; the study of the canarian indigenous people and of the people who came there at different historical moments. Furthermore, it is necessary to know the rediscovery of the Canary islands and their processes of conquest and evangelization. All these data will offer a new and surprising outlook to understanding not only the letters but the same wood sculpture. From this moment the inquiry will move to see if the letters correspond to any possible language, which has led us to analyze a very large set of texts as close to the time under study, in a hundred languages. After the language examination, has been carried out a study of possible cryptographic forms used to generate the text of the inscription. A detailed and thorough list of possible techniques that could be adopted is offered. Then exhaustively we cryptanalyze the letters of the marian carving. At the same time a new crypto framework of methods and systems more orderly and complete, especially since the emergence of asymmetric key cryptography, is provided. We continue researching the possible pseudo-random generation of the text, a text that would not make any sense. At this time, and having completed all the possibilities and hypotheses, all having refused, we return to rethink our assumptions. From there we analyze in depth the ethnographic and linguistic berber sphere as the most likely hypothesis. Following the deepening of this language and correcting the mistakes that led us not to detect it in our analysis above, we conclude that this is an archaic berber language, a set of letters belonging to a language and family not extinct today but very modeled and influenced by other languages, primarily arabic. This will lead us to rescue lexical, morphological, syntactic and phonetic aspects of this archaic speech. With all this data we make a wide semantic study of the carving from the indigenous and christian perspective. Finally, from the lexical voices and roots of the berber languages amazigh and island-amazigh, we give the meaning of the letters inscribed in the marian carving of Candelaria.

Brief Announcement: Node Sampling Using Centrifugal Random Walks.

We propose distributed algorithms for sampling networks based on a new class of random walks that we call Centrifugal Random Walks (CRW). A CRW is a random walk that starts at a source and always moves away from it. We propose CRW algorithms for connected networks with arbitrary probability distributions, and for grids and networks with regular concentric connectivity with distance based distributions. All CRW sampling algorithms select a node with the exact probability distribution, do not need warm-up, and end in a number of hops bounded by the network diameter.

Node sampling using random centrifugal walks

Sampling a network with a given probability distribution has been identified as a useful operation. In this paper we propose distributed algorithms for sampling networks, so that nodes are selected by a special node, called the source, with a given probability distribution. All these algorithms are based on a new class of random walks, that we call Random Centrifugal Walks (RCW). A RCW is a random walk that starts at the source and always moves away from it. Firstly, an algorithm to sample any connected network using RCW is proposed. The algorithm assumes that each node has a weight, so that the sampling process must select a node with a probability proportional to its weight. This algorithm requires a preprocessing phase before the sampling of nodes. In particular, a minimum diameter spanning tree (MDST) is created in the network, and then nodes weights are efficiently aggregated using the tree. The good news are that the preprocessing is done only once, regardless of the number of sources and the number of samples taken from the network. After that, every sample is done with a RCW whose length is bounded by the network diameter. Secondly, RCW algorithms that do not require preprocessing are proposed for grids and networks with regular concentric connectivity, for the case when the probability of selecting a node is a function of its distance to the source. The key features of the RCW algorithms (unlike previous Markovian approaches) are that (1) they do not need to warm-up (stabilize), (2) the sampling always finishes in a number of hops bounded by the network diameter, and (3) it selects a node with the exact probability distribution.

Modelización de series temporales complejas: de las redes de telecomunicación a los mercados financieros

Este trabajo aborda el problema de modelizar sistemas din´amicos reales a partir del estudio de sus series temporales, usando una formulaci´on est´andar que pretende ser una abstracci´on universal de los sistemas din´amicos, independientemente de su naturaleza determinista, estoc´astica o h´ıbrida. Se parte de modelizaciones separadas de sistemas deterministas por un lado y estoc´asticos por otro, para converger finalmente en un modelo h´ıbrido que permite estudiar sistemas gen´ericos mixtos, esto es, que presentan una combinaci´on de comportamiento determinista y aleatorio. Este modelo consta de dos componentes, uno determinista consistente en una ecuaci´on en diferencias, obtenida a partir de un estudio de autocorrelaci´on, y otro estoc´astico que modeliza el error cometido por el primero. El componente estoc´astico es un generador universal de distribuciones de probabilidad, basado en un proceso compuesto de variables aleatorias, uniformemente distribuidas en un intervalo variable en el tiempo. Este generador universal es deducido en la tesis a partir de una nueva teor´ıa sobre la oferta y la demanda de un recurso gen´erico. El modelo resultante puede formularse conceptualmente como una entidad con tres elementos fundamentales: un motor generador de din´amica determinista, una fuente interna de ruido generadora de incertidumbre y una exposici´on al entorno que representa las interacciones del sistema real con el mundo exterior. En las aplicaciones estos tres elementos se ajustan en base al hist´orico de las series temporales del sistema din´amico. Una vez ajustados sus componentes, el modelo se comporta de una forma adaptativa tomando como inputs los nuevos valores de las series temporales del sistema y calculando predicciones sobre su comportamiento futuro. Cada predicci´on se presenta como un intervalo dentro del cual cualquier valor es equipro- bable, teniendo probabilidad nula cualquier valor externo al intervalo. De esta forma el modelo computa el comportamiento futuro y su nivel de incertidumbre en base al estado actual del sistema. Se ha aplicado el modelo en esta tesis a sistemas muy diferentes mostrando ser muy flexible para afrontar el estudio de campos de naturaleza dispar. El intercambio de tr´afico telef´onico entre operadores de telefon´ıa, la evoluci´on de mercados financieros y el flujo de informaci´on entre servidores de Internet son estudiados en profundidad en la tesis. Todos estos sistemas son modelizados de forma exitosa con un mismo lenguaje, a pesar de tratarse de sistemas f´ısicos totalmente distintos. El estudio de las redes de telefon´ıa muestra que los patrones de tr´afico telef´onico presentan una fuerte pseudo-periodicidad semanal contaminada con una gran cantidad de ruido, sobre todo en el caso de llamadas internacionales. El estudio de los mercados financieros muestra por su parte que la naturaleza fundamental de ´estos es aleatoria con un rango de comportamiento relativamente acotado. Una parte de la tesis se dedica a explicar algunas de las manifestaciones emp´ıricas m´as importantes en los mercados financieros como son los “fat tails”, “power laws” y “volatility clustering”. Por ´ultimo se demuestra que la comunicaci´on entre servidores de Internet tiene, al igual que los mercados financieros, una componente subyacente totalmente estoc´astica pero de comportamiento bastante “d´ocil”, siendo esta docilidad m´as acusada a medida que aumenta la distancia entre servidores. Dos aspectos son destacables en el modelo, su adaptabilidad y su universalidad. El primero es debido a que, una vez ajustados los par´ametros generales, el modelo se “alimenta” de los valores observables del sistema y es capaz de calcular con ellos comportamientos futuros. A pesar de tener unos par´ametros fijos, la variabilidad en los observables que sirven de input al modelo llevan a una gran riqueza de ouputs posibles. El segundo aspecto se debe a la formulaci´on gen´erica del modelo h´ıbrido y a que sus par´ametros se ajustan en base a manifestaciones externas del sistema en estudio, y no en base a sus caracter´ısticas f´ısicas. Estos factores hacen que el modelo pueda utilizarse en gran variedad de campos. Por ´ultimo, la tesis propone en su parte final otros campos donde se han obtenido ´exitos preliminares muy prometedores como son la modelizaci´on del riesgo financiero, los algoritmos de routing en redes de telecomunicaci´on y el cambio clim´atico. Abstract This work faces the problem of modeling dynamical systems based on the study of its time series, by using a standard language that aims to be an universal abstraction of dynamical systems, irrespective of their deterministic, stochastic or hybrid nature. Deterministic and stochastic models are developed separately to be merged subsequently into a hybrid model, which allows the study of generic systems, that is to say, those having both deterministic and random behavior. This model is a combination of two different components. One of them is deterministic and consisting in an equation in differences derived from an auto-correlation study and the other is stochastic and models the errors made by the deterministic one. The stochastic component is an universal generator of probability distributions based on a process consisting in random variables distributed uniformly within an interval varying in time. This universal generator is derived in the thesis from a new theory of offer and demand for a generic resource. The resulting model can be visualized as an entity with three fundamental elements: an engine generating deterministic dynamics, an internal source of noise generating uncertainty and an exposure to the environment which depicts the interactions between the real system and the external world. In the applications these three elements are adjusted to the history of the time series from the dynamical system. Once its components have been adjusted, the model behaves in an adaptive way by using the new time series values from the system as inputs and calculating predictions about its future behavior. Every prediction is provided as an interval, where any inner value is equally probable while all outer ones have null probability. So, the model computes the future behavior and its level of uncertainty based on the current state of the system. The model is applied to quite different systems in this thesis, showing to be very flexible when facing the study of fields with diverse nature. The exchange of traffic between telephony operators, the evolution of financial markets and the flow of information between servers on the Internet are deeply studied in this thesis. All these systems are successfully modeled by using the same “language”, in spite the fact that they are systems physically radically different. The study of telephony networks shows that the traffic patterns are strongly weekly pseudo-periodic but mixed with a great amount of noise, specially in the case of international calls. It is proved that the underlying nature of financial markets is random with a moderate range of variability. A part of this thesis is devoted to explain some of the most important empirical observations in financial markets, such as “fat tails”, “power laws” and “volatility clustering”. Finally it is proved that the communication between two servers on the Internet has, as in the case of financial markets, an underlaying random dynamics but with a narrow range of variability, being this lack of variability more marked as the distance between servers is increased. Two aspects of the model stand out as being the most important: its adaptability and its universality. The first one is due to the fact that once the general parameters have been adjusted , the model is “fed” on the observable manifestations of the system in order to calculate its future behavior. Despite the fact that the model has fixed parameters the variability in the observable manifestations of the system, which are used as inputs of the model, lead to a great variability in the possible outputs. The second aspect is due to the general “language” used in the formulation of the hybrid model and to the fact that its parameters are adjusted based on external manifestations of the system under study instead of its physical characteristics. These factors made the model suitable to be used in great variety of fields. Lastly, this thesis proposes other fields in which preliminary and promising results have been obtained, such as the modeling of financial risk, the development of routing algorithms for telecommunication networks and the assessment of climate change.

Novel fast random search clustering approach for mixing matrix identification in MIMO linear blind inverse problems with sparse inputs

In this paper we propose a novel fast random search clustering (RSC) algorithm for mixing matrix identification in multiple input multiple output (MIMO) linear blind inverse problems with sparse inputs. The proposed approach is based on the clustering of the observations around the directions given by the columns of the mixing matrix that occurs typically for sparse inputs. Exploiting this fact, the RSC algorithm proceeds by parameterizing the mixing matrix using hyperspherical coordinates, randomly selecting candidate basis vectors (i.e. clustering directions) from the observations, and accepting or rejecting them according to a binary hypothesis test based on the Neyman–Pearson criterion. The RSC algorithm is not tailored to any specific distribution for the sources, can deal with an arbitrary number of inputs and outputs (thus solving the difficult under-determined problem), and is applicable to both instantaneous and convolutive mixtures. Extensive simulations for synthetic and real data with different number of inputs and outputs, data size, sparsity factors of the inputs and signal to noise ratios confirm the good performance of the proposed approach under moderate/high signal to noise ratios. RESUMEN. Método de separación ciega de fuentes para señales dispersas basado en la identificación de la matriz de mezcla mediante técnicas de "clustering" aleatorio.

Nonparametric generalized belief propagation based on pseudo-junction tree for cooperative localization in wireless networks

Non-parametric belief propagation (NBP) is a well-known message passing method for cooperative localization in wireless networks. However, due to the over-counting problem in the networks with loops, NBP’s convergence is not guaranteed, and its estimates are typically less accurate. One solution for this problem is non-parametric generalized belief propagation based on junction tree. However, this method is intractable in large-scale networks due to the high-complexity of the junction tree formation, and the high-dimensionality of the particles. Therefore, in this article, we propose the non-parametric generalized belief propagation based on pseudo-junction tree (NGBP-PJT). The main difference comparing with the standard method is the formation of pseudo-junction tree, which represents the approximated junction tree based on thin graph. In addition, in order to decrease the number of high-dimensional particles, we use more informative importance density function, and reduce the dimensionality of the messages. As by-product, we also propose NBP based on thin graph (NBP-TG), a cheaper variant of NBP, which runs on the same graph as NGBP-PJT. According to our simulation and experimental results, NGBP-PJT method outperforms NBP and NBP-TG in terms of accuracy, computational, and communication cost in reasonably sized networks.

Phase transitions in number theory: from the birthday problem to Sidon sets

In this work, we show how number theoretical problems can be fruitfully approached with the tools of statistical physics. We focus on g-Sidon sets, which describe sequences of integers whose pairwise sums are different, and propose a random decision problem which addresses the probability of a random set of k integers to be g-Sidon. First, we provide numerical evidence showing that there is a crossover between satisfiable and unsatisfiable phases which converts to an abrupt phase transition in a properly defined thermodynamic limit. Initially assuming independence, we then develop a mean-field theory for the g-Sidon decision problem. We further improve the mean-field theory, which is only qualitatively correct, by incorporating deviations from independence, yielding results in good quantitative agreement with the numerics for both finite systems and in the thermodynamic limit. Connections between the generalized birthday problem in probability theory, the number theory of Sidon sets and the properties of q-Potts models in condensed matter physics are briefly discussed

Computer simulation of random parckings for self-similar particle size distributions in soil and granular materials: porosity and pore size distribution

A 2D computer simulation method of random packings is applied to sets of particles generated by a self-similar uniparametric model for particle size distributions (PSDs) in granular media. The parameter p which controls the model is the proportion of mass of particles corresponding to the left half of the normalized size interval [0,1]. First the influence on the total porosity of the parameter p is analyzed and interpreted. It is shown that such parameter, and the fractal exponent of the associated power scaling, are efficient packing parameters, but this last one is not in the way predicted in a former published work addressing an analogous research in artificial granular materials. The total porosity reaches the minimum value for p = 0.6. Limited information on the pore size distribution is obtained from the packing simulations and by means of morphological analysis methods. Results show that the range of pore sizes increases for decreasing values of p showing also different shape in the volume pore size distribution. Further research including simulations with a greater number of particles and image resolution are required to obtain finer results on the hierarchical structure of pore space.

Arquitectura industrial tabacalera en la España peninsular: secaderos y fábricas

La Arquitectura industrial del tabaco en España está representada por dos tipos de construcciones, que corresponden a las dos fases en las que se divide el proceso de producción del tabaco: los secaderos (arquitectura bioclimática donde se realiza el secado), y las fábricas (centros donde se elabora el tabaco procedente de los secaderos). Las fábricas se repartieron por todo el territorio español, ocupando preferiblemente los lugares costeros, aunque existen casos en los que su localización obedecía a razones políticas. Estos edificios, en su mayoría, incluidos en los centros históricos de las ciudades, han cambiando de uso, y las antiguas fábricas de tabaco se han transformado en su mayoría, en centros de cultura, o centros sociales y representativos. La tesis surge del análisis de las características constructivas de la arquitectura industrial del tabaco: de los secaderos y de las fábricas, por tratarse de una tipología con suficiente entidad y un ejemplo de arquitectura bioclimática de producción en el caso de los secaderos, y por conseguir haberse adaptado a otros usos en el caso de las fábricas. La arquitectura de producción emplea un lenguaje acorde con los avances de la industrialización, anticipando materiales y estructuras, y condensando en una tipología específica las complejas relaciones establecidas entre producto, hombres y espacio. Estos edificios tuvieron una extensa implantación en el territorio, y se caracterizan por una serie de valores tecnológicos, arquitectónicos, sociológicos y paisajísticos, que hacen de ellos un documento de primera magnitud para conocer: la evolución e implantación de las técnicas constructivas (materiales y estructuras), los procesos de innovación tipológica y la estructura económica y procedimientos técnicos utilizados. El territorio en el que se insertan constituye su contexto territorial, por lo que no sería adecuado considerar estos edificios como elementos aislados, sin analizar la relación con el entorno en el que se generaron. Por este motivo, se analizan las condiciones higrotérmicas ambientales de los secaderos para compararlas con las de confort humano y establecer relaciones y parámetros compatibles. Los ejemplos analizados de secaderos son todos de fábrica. El uso del ladrillo como módulo principal para la elaboración de un edificio, supone la consideración de un “grado cero” de todo el aparato constructivo y compositivo de la arquitectura. Dejar el ladrillo visto, supone hacer explícitos todos los procesos acumulativos. Este elemento mínimo, permite unas posibilidades enormemente abiertas, pero no absolutamente aleatorias, que definen su propia lógica combinatoria. La exigencia de sinceridad, característica de la arquitectura industrial, en la exposición de los materiales, exhibiéndolos en su propia naturaleza y en el modo real de ser utilizados, se hace patente en este tipo de construcción. Se realiza un estudio de permeabilidad en las fachadas de los secaderos, para determinar el grado de ventilación y su relación con la orientación, el patrón de celosía empleado y el volumen total. Este sistema de acondicionamiento climático específico, puede servir de recurso a otras construcciones, por lo que se podría trasladar el sistema constructivo y formal de los secaderos a otros usos, desde una doble vertiente: Arquitectura para la adaptación climática al entorno. Arquitectura como generadora de condiciones climáticas específicas, en el interior. La utilidad de los secaderos es fundamentalmente: proporcionar sombra, ventilación y un espacio cubierto, pero permeable en sus fachadas. La arquitectura industrial debe ser reconocida dentro del conjunto patrimonial, debido a sus características propias que permiten su diferenciación del resto de la arquitectura. Conocer la estructura productiva permite analizar correctamente estas construcciones, ya que el programa inicial es básico para entender la organización del espacio interior. Las fábricas no se situaron cerca de las zonas de producción del tabaco, excepto en dos casos: Cádiz y Palazuelo, en los que existen secaderos y campos de cultivo de hoja de tabaco en las áreas cercanas. La principal causa de esta separación es que el proceso de obtención de tabaco es un proceso dividido en dos fases principales: proceso primario y proceso secundario. En el proceso primario la hoja de tabaco se seca en los secaderos, en los que es determinante el clima, pero únicamente en el caso del secado del tabaco al aire. En el proceso secundario sin embargo, el tabaco llega previamente tratado a las fábricas, por lo que no influye el clima en esta parte del proceso. Esta razón determina que en las áreas climáticas donde se centra el estudio, haya zonas en las que existen fábrica y secaderos y otras en las que únicamente existe fábrica, o sólo secaderos. La localización de las fábricas atendía a razones de muy diferente índole, las más importantes fueron: geográficas, estratégicas, y políticas. En la mayoría de las fábricas la elección de la ciudad de emplazamiento estaba ligada a la recepción de la materia prima, que principalmente se hacía por vía marítima, o acuática (el caso de Sevilla), y por vía terrestre, utilizando como medio de transporte el ferrocarril. Sólo dos casos, de las antiguas fábricas, corresponden a razones políticas, son las dos únicas que no están en la costa: Madrid y Logroño. La de Madrid se construyó por centralidad política, y porque geográficamente ocupaba el punto central de todas las comunicaciones terrestres por carretera y ferrocarril. Muchas de las fábricas se situaron cercanas a las estaciones de ferrocarril. La de Logroño atendió, sin embargo, a razones políticas. Para finalizar, se realiza un estudio comparativo de las fábricas de Sevilla, Madrid y San Sebastián. Las razones que justifican esta elección son: - La de Sevilla fue históricamente la primera fábrica y la más importante. - La de Madrid fue la más importante a nivel administrativo, la sede de Tabacalera se instaló en la capital, y después de la de Sevilla, fue la que sirvió de modelo al resto de las fábricas. - La de San Sebastián era la más grande del Norte. Los análisis que se han realizado son de: volumen y superficies de patios, superficies de cubierta, permeabilidad o huecos en fachadas, orientación y soleamiento de patios, distribución espacial interior y organización, y evolución de usos. Podemos observar que en la mayoría de estas fábricas ha habido una transformación en el uso, pasando de ser edificios industriales a edificios culturales. Estas construcciones se pueden considerar como infraestructuras adaptables, por ser útiles, sostenibles y funcionales. ABSTRACT The Spanish industrial architecture of tobacco is represented by two construction types that correspond to the two phases of tobacco production: the drying sheds (bioclimatic constructions where the drying process takes place) and factories (centres where tobacco is processed after the drying process). The factories were distributed throughout the Spanish territory, preferably occupying coastal locations, although some of them were located elsewhere following political reasons. Most of the buildings inside city centres have suffered changes in their use, becoming cultural, social or representative centres. This thesis attempts the analysis of the constructive systems employed in tobacco industrial architecture, from drying sheds to factories. The drying sheds are an example of bioclimatic industrial architecture. The factories are a typology that have successfully adapted to new uses. Industrial architecture uses a language that follows the advances in industrialization, anticipating new materials and structures, and merging the complex relationships established among products, human beings, space and locations. These buildings were promoted extensively in the country. They are characterized by technological architectural sociological and landscaping innovations. They are considered as important examples of the evolution and the implementation of construction techniques (building materials and structures). They are also considered as examples of innovation in the building typology, in their economic structure and in the technologies that they have applied. The settings in which the drying sheds are placed have an important influence in them. They cannot be considered as isolated elements. Instead, there is a close relationship between drying sheds and the surroundings in which they are located. Drying sheds’ hygrotermal and environmental conditions are analyzed in this thesis to compare them with the values of human comfort and find suitable relationships and parameters. All the drying sheds that have been analyzed are constructed with brick. This implies a consideration of “zero degree” for both the construction and the composition of the architectural process. The detailing - entails making all the accumulative processes explicit as the brick walls are left exposed. This minimal component allows a wide range of options that are never random but based on the logic of the way in which it is combined. The “sincerity” in the exposition of material, displaying them in their very nature and showing how they are really used, is a basic characteristic of industrial architecture, and it is even more expressive in these types of buildings. The walls of the drying sheds undergo a permeability assessment in order to determine the degree of ventilation and orientation, the lattice pattern used and the overall volume. This specific conditioning system can serve as a resource for other buildings, and consequently, it could be transferred to other uses within a two-pronged approach: -Climatically adapted architecture that takes into account the surroundings. -Architecture as a generator of specific climatic conditions indoors. Drying sheds’ main purposes / aims deal with how to provide shade, ventilation and a covered space as well as permeability. The industrial architecture must be recognized as historical valuable buildings due to its intrinsic and distinctive characteristics. Knowing the productive structure, allow us to make a proper analysis of these buildings, since the basic aim, is to understand the spatial organization indoors. Factories did not come close to the tobacco production, with the exception of Cádiz and Palazuelo, where there are sheds and tobacco croplands nearby. The main reason for this separation is that the process of obtaining tobacco has two processes: the primary process and the secondary process. In the primary process tobacco leaves are left to dry. In the secondary process, previously manufactured tobacco allocated in the factories where the weather conditions are not important. This fact determines that in the climate areas where this study tales place there are some cases in which we can find both factories and drying sheds, and others where there are either factories or drying sheds only. The location of these factories met various demands, being the most outstanding the ones related to geographic, strategic and political reasons. In most factories the choice of its location was often linked to the incoming of raw goods, mainly delivered through waterways –it is the case of Seville,) and by land, using railways. The location of the factories was linked to political reasons in only two cases Madrid and Logroño, which are the only ones that are not placed near the coast. The one in Madrid was built due to its political centrality and because geographically speaking, it was the reference landmark of means of land and rail transports. Many factories, in fact, were settled nearby rail stations. For the factory in Logroño, only political reasons were taken into consideration. I should like to close by undertaking a comparative study of factories in Seville, Madrid and San Sebastian. There are a number of reasons to substantiate this choice: -The factory in Seville was historically speaking the first that was built and the most important one. -The factory in Madrid was the most important one administratively. This factory was the headquarters as well as being, after Seville, the one which provided a model for other factories. -The factory in San Sebastian is the biggest in the North of Spain. The analysis carried out are related to the volume of the buildings and the surface areas of the courtyards, the surface of the roofs, the permeability of the walls and the openings of the façade, the orientation and the sun exposure, the indoor spatial distribution and organization and evolution of the uses (formerly and currently) I observe that in most of these factories there has been a change in the use of the buildings, from industrial cultural purposes. These buildings can be considered as adaptable infrastructures based on a combination of architectural practicability, sustainability and functionality.