Asymptotically Optimum Estimation of a Probability in Inverse Binomial Sampling under General Loss Functions

The optimum quality that can be asymptotically achieved in the estimation of a probability p using inverse binomial sampling is addressed. A general definition of quality is used in terms of the risk associated with a loss function that satisfies certain assumptions. It is shown that the limit superior of the risk for p asymptotically small has a minimum over all (possibly randomized) estimators. This minimum is achieved by certain non-randomized estimators. The model includes commonly used quality criteria as particular cases. Applications to the non-asymptotic regime are discussed considering specific loss functions, for which minimax estimators are derived.

Video analysis based vehicle detection and tracking using an MCMC sampling framework

This article presents a probabilistic method for vehicle detection and tracking through the analysis of monocular images obtained from a vehicle-mounted camera. The method is designed to address the main shortcomings of traditional particle filtering approaches, namely Bayesian methods based on importance sampling, for use in traffic environments. These methods do not scale well when the dimensionality of the feature space grows, which creates significant limitations when tracking multiple objects. Alternatively, the proposed method is based on a Markov chain Monte Carlo (MCMC) approach, which allows efficient sampling of the feature space. The method involves important contributions in both the motion and the observation models of the tracker. Indeed, as opposed to particle filter-based tracking methods in the literature, which typically resort to observation models based on appearance or template matching, in this study a likelihood model that combines appearance analysis with information from motion parallax is introduced. Regarding the motion model, a new interaction treatment is defined based on Markov random fields (MRF) that allows for the handling of possible inter-dependencies in vehicle trajectories. As for vehicle detection, the method relies on a supervised classification stage using support vector machines (SVM). The contribution in this field is twofold. First, a new descriptor based on the analysis of gradient orientations in concentric rectangles is dened. This descriptor involves a much smaller feature space compared to traditional descriptors, which are too costly for real-time applications. Second, a new vehicle image database is generated to train the SVM and made public. The proposed vehicle detection and tracking method is proven to outperform existing methods and to successfully handle challenging situations in the test sequences.

A matrix pencil approach to the existence of compactly supported reconstruction functions in average sampling

The aim of this work is to solve a question raised for average sampling in shift-invariant spaces by using the well-known matrix pencil theory. In many common situations in sampling theory, the available data are samples of some convolution operator acting on the function itself: this leads to the problem of average sampling, also known as generalized sampling. In this paper we deal with the existence of a sampling formula involving these samples and having reconstruction functions with compact support. Thus, low computational complexity is involved and truncation errors are avoided. In practice, it is accomplished by means of a FIR filter bank. An answer is given in the light of the generalized sampling theory by using the oversampling technique: more samples than strictly necessary are used. The original problem reduces to finding a polynomial left inverse of a polynomial matrix intimately related to the sampling problem which, for a suitable choice of the sampling period, becomes a matrix pencil. This matrix pencil approach allows us to obtain a practical method for computing the compactly supported reconstruction functions for the important case where the oversampling rate is minimum. Moreover, the optimality of the obtained solution is established.

Sampling strategies using the "accumulation chamber" for monitoring geological

Fundación Ciudad de la Energía (CIUDEN) is carrying out a project of geological storage of CO2, where CO2 injection tests are planned in saline aquifers at a depth of 1500 m for scientific objectives and project demonstration. Before any CO2 is stored, it is necessary to determine the baseline flux of CO2 in order to detect potential leakage during injection and post-injection monitoring. In November 2009 diffuse flux measurements of CO2 using an accumulationchamber were made in the area selected by CIUDEN for geological storage, located in Hontomin province of Burgos (Spain). This paper presents the tests carried out in order to establish the optimum sampling methodology and the geostatistical analyses performed to determine the range, with which future field campaigns will be planned.

Simultaneous 3D object tracking and camera parameter estimation by Bayesian methods and transdimensional MCMC sampling

Multi-camera 3D tracking systems with overlapping cameras represent a powerful mean for scene analysis, as they potentially allow greater robustness than monocular systems and provide useful 3D information about object location and movement. However, their performance relies on accurately calibrated camera networks, which is not a realistic assumption in real surveillance environments. Here, we introduce a multi-camera system for tracking the 3D position of a varying number of objects and simultaneously refin-ing the calibration of the network of overlapping cameras. Therefore, we introduce a Bayesian framework that combines Particle Filtering for tracking with recursive Bayesian estimation methods by means of adapted transdimensional MCMC sampling. Addi-tionally, the system has been designed to work on simple motion detection masks, making it suitable for camera networks with low transmission capabilities. Tests show that our approach allows a successful performance even when starting from clearly inaccurate camera calibrations, which would ruin conventional approaches.

Coevolución virus-planta: impacto de las infecciones virales en poblaciones silvestres de Chiltepín (capsicum annuum var. Aviculare (dierbach) d’arcy & eshbaugh) en México.

El impacto negativo que tienen los virus en las plantas hace que estos puedan ejercer un papel ecológico como moduladores de la dinámica espacio-temporal de las poblaciones de sus huéspedes. Entender cuáles son los mecanismos genéticos y los factores ambientales que determinan tanto la epidemiología como la estructura genética de las poblaciones de virus puede resultar de gran ayuda para la comprensión del papel ecológico de las infecciones virales. Sin embargo, existen pocos trabajos experimentales que hayan abordado esta cuestión. En esta tesis, se analiza el efecto de la heterogeneidad del paisaje sobre la incidencia de los virus y la estructura genética de sus poblaciones. Asimismo, se explora como dichos factores ambientales influyen en la importancia relativa que los principales mecanismos de generación de variabilidad genética (mutación, recombinación y migración) tienen en la evolución de los virus. Para ello se ha usado como sistema los begomovirus que infectan poblaciones de chiltepín (Capsicum annuum var. aviculare (Dierbach) D´Arcy & Eshbaugh) en México. Se analizó la incidencia de diferentes virus en poblaciones de chiltepín distribuidas a lo largo de seis provincias biogeográficas, representando el área de distribución de la especie en México, y localizadas en hábitats con diferente grado de intervención humana: poblaciones sin intervención humana (silvestres); poblaciones toleradas (lindes y pastizales), y poblaciones manejadas por el hombre (monocultivos y huertos familiares). Entre los virus analizados, los begomovirus mostraron la mayor incidencia, detectándose en todas las poblaciones y años de muestreo. Las únicas dos especies de begomovirus que se encontraron infectando al chiltepín fueron: el virus del mosaico dorado del chile (Pepper golden mosaic virus, PepGMV) y el virus huasteco del amarilleo de venas del chile (Pepper huasteco yellow vein virus, PHYVV). Por ello, todos los análisis realizados en esta tesis se centran en estas dos especies de virus. La incidencia de PepGMV y PHYVV, tanto en infecciones simples como mixtas, aumento cuanto mayor fue el nivel de intervención humana en las poblaciones de chiltepín, lo que a su vez se asoció con una menor biodiversidad y una mayor densidad de plantas. Además, la incidencia de infecciones mixtas, altamente relacionada con la presencia de síntomas, fue también mayor en las poblaciones cultivadas. La incidencia de estos dos virus también varió en función de la población de chiltepín y de la provincia biogeográfica. Por tanto, estos resultados apoyan una de las hipótesis XVI clásicas de la Patología Vegetal según la cual la simplificación de los ecosistemas naturales debida a la intervención humana conduce a un mayor riesgo de enfermedad de las plantas, e ilustran sobre la importancia de la heterogeneidad del paisaje a diferentes escalas en la determinación de patrones epidemiológicos. La heterogeneidad del paisaje no solo afectó a la epidemiología de PepGMV y PHYVV, sino también a la estructura genética de sus poblaciones. En ambos virus, el nivel de diferenciación genética mayor fue la población, probablemente asociado a la capacidad de migración de su vector Bemisia tabaci; y en segundo lugar la provincia biogeográfica, lo que podría estar relacionado con el papel del ser humano como agente dispersor de PepGMV y PHYVV. La estima de las tasas de sustitución nucleotídica de las poblaciones de PepGMV y PHYVV mostró una rápida dinámica evolutiva. Los árboles filogenéticos de ambos virus presentaron una topología en estrella, lo que sugiere una expansión reciente en las poblaciones de chiltepín. La reconstrucción de los patrones de migración de ambos virus indicó que ésta expansión parece haberse producido desde la zona central de México siguiendo un patrón radial, y en los últimos 30 años. Es importante tener en cuenta que el patrón espacial de la diversidad genética de las poblaciones de PepGMV y PHYVV es similar al descrito previamente para el chiltepín lo que podría dar lugar a la congruencia de las genealogías del huésped y la de los virus. Dicha congruencia se encontró cuando se tuvieron en cuenta únicamente las poblaciones de hábitats silvestres y tolerados, lo que probablemente se debe a una codivergencia en el espacio pero no en el tiempo, dado que la evolución de virus y huésped han ocurrido a escalas temporales muy diferentes. Finalmente, el análisis de la frecuencia de recombinación en PepGMV y PHYVV indicó que esta juega un papel importante en la evolución de ambos virus, dependiendo su importancia del nivel de intervención humana de la población de chiltepín. Este factor afectó también a la intensidad de la selección a la que se ven sometidos los genomas de PepGMV y PHYVV. Los resultados de esta tesis ponen de manifiesto la importancia que la reducción de la biodiversidad asociada al nivel de intervención humana de las poblaciones de plantas y la heterogeneidad del paisaje tiene en la emergencia de nuevas enfermedades virales. Por tanto, es necesario considerar estos factores ambientales a la hora de comprender la epidemiologia y la evolución de los virus de plantas.XVII SUMMARY Plant viruses play a key role as modulators of the spatio-temporal dynamics of their host populations, due to their negative impact in plant fitness. Knowledge on the genetic and environmental factors that determine the epidemiology and the genetic structure of virus populations may help to understand the ecological role of viral infections. However, few experimental works have addressed this issue. This thesis analyses the effect of landscape heterogeneity in the prevalence of viruses and the genetic structure of their populations. Also, how these environmental factors influence the relative importance of the main mechanisms for generating genetic variability (mutation, recombination and migration) during virus evolution is explored. To do so, the begomoviruses infecting chiltepin (Capsicum annuum var. aviculare (Dierbach) D'Arcy & Eshbaugh) populations in Mexico were used. Incidence of different viruses in chiltepin populations of six biogeographical provinces representing the species distribution in Mexico was determined. Populations belonged to different habitats according to the level of human management: populations with no human intervention (Wild); populations naturally dispersed and tolerated in managed habitats (let-standing), and human managed populations (cultivated). Among the analyzed viruses, the begomoviruses showed the highest prevalence, being detected in all populations and sampling years. Only two begomovirus species infected chiltepin: Pepper golden mosaic virus, PepGMV and Pepper huasteco yellow vein virus, PHYVV. Therefore, all the analyses presented in this thesis are focused in these two viruses. The prevalence of PepGMV and PHYVV, in single and mixed infections, increased with higher levels of human management of the host population, which was associated with decreased biodiversity and increased plant density. Furthermore, cultivated populations showed higher prevalence of mixed infections and symptomatic plants. The prevalence of the two viruses also varied depending on the chiltepin population and on the biogeographical province. Therefore, these results support a classical hypothesis of Plant Pathology stating that simplification of natural ecosystems due to human management leads to an increased disease risk, and illustrate on the importance of landscape heterogeneity in determining epidemiological patterns. Landscape heterogeneity not only affected the epidemiology of PepGMV and PHYVV, but also the genetic structure of their populations. Both viruses had the highest level of genetic differentiation at the population scale, probably associated with the XVIII migration patterns of its vector Bemisia tabaci, and a second level at the biogeographical province scale, which could be related to the role of humans as dispersal agents of PepGMV and PHYVV. The estimates of nucleotide substitution rates of the virus populations indicated rapid evolutionary dynamics. Accordingly, phylogenetic trees of both viruses showed a star topology, suggesting a recent diversification in the chiltepin populations. Reconstruction of PepGMV and PHYVV migration patterns indicated that they expanded from central Mexico following a radial pattern during the last 30 years. Importantly, the spatial genetic structures of the virus populations were similar to that described previously for the chiltepin, which may result in the congruence of the host and virus genealogies. Such congruence was found only in wild and let-standing populations. This is probably due to a co-divergence in space but not in time, given the different evolutionary time scales of the host and virus populations. Finally, the frequency of recombination detected in the PepGMV and PHYVV populations indicated that this mechanism plays an important role in the evolution of both viruses at the intra-specific scale. The level of human management had a minor effect on the frequency of recombination, but influenced the strength of negative selective pressures in the viral genomes. The results of this thesis highlight the importance of decreased biodiversity in plant populations associated with the level of human management and of landscape heterogeneity on the emergence of new viral diseases. Therefore it is necessary to consider these environmental factors in order to fully understand the epidemiology and evolution of plant viruses.

Morphological and germination response variability in seeds of wild yellow gentian (Gentiana lutea L.) accessions from northwest Spain

Gentiana lutea L. (yellow gentian, Gentianaceae) is an important medicinal plant under protection as endangered species in most European countries. The aim of this work was to evaluate variation in seed mass, seed water content, and seed germination among 56 wild accessions of G. lutea. The effect of gibberellic acid (GA3), putrescine, moist chilling, and level of ripeness of seeds on subsequent germination was also investigated. Seeds of G. lutea showed physiological dormancy (final germination percentages ranged from 0% to 11%, depending on the accession) and GA3 enhanced seed germination drastically in all the accessions. The highest germination (99%) of GA3-treated seeds was reached at 15 °C. Final germination percentage and germination rate (as expressed by mean germination time), as well as seed mass and seed water content, varied significantly among accessions. In general, 1 year moist chilling did not significantly enhance G. lutea seed germination. For most accessions, no significant differences were found between fully ripe seeds and less ripe seeds for seed water content, seed mass, and seed germination. Applications of GA3 were always most effective than those of putrescine for increasing seed germination.

Estimation of a probability in inverse binomial sampling under normalized linear-linear and inverse-linear loss

Sequential estimation of the success probability p in inverse binomial sampling is considered in this paper. For any estimator pˆ , its quality is measured by the risk associated with normalized loss functions of linear-linear or inverse-linear form. These functions are possibly asymmetric, with arbitrary slope parameters a and b for pˆ

p , respectively. Interest in these functions is motivated by their significance and potential uses, which are briefly discussed. Estimators are given for which the risk has an asymptotic value as p→0, and which guarantee that, for any p∈(0,1), the risk is lower than its asymptotic value. This allows selecting the required number of successes, r, to meet a prescribed quality irrespective of the unknown p. In addition, the proposed estimators are shown to be approximately minimax when a/b does not deviate too much from 1, and asymptotically minimax as r→∞ when a=b.

Analysis and design of multirate synchronous sampling schemes for sparse multiband signals

We consider the problem of developing efficient sampling schemes for multiband sparse signals. Previous results on multicoset sampling implementations that lead to universal sampling patterns (which guarantee perfect reconstruction), are based on a set of appropriate interleaved analog to digital converters, all of them operating at the same sampling frequency. In this paper we propose an alternative multirate synchronous implementation of multicoset codes, that is, all the analog to digital converters in the sampling scheme operate at different sampling frequencies, without need of introducing any delay. The interleaving is achieved through the usage of different rates, whose sum is significantly lower than the Nyquist rate of the multiband signal. To obtain universal patterns the sampling matrix is formulated and analyzed. Appropriate choices of the parameters, that is the block length and the sampling rates, are also proposed.

Design of universal multicoset sampling patterns for compressed sensing of multiband sparse signals

Many problems in digital communications involve wideband radio signals. As the most recent example, the impressive advances in Cognitive Radio systems make even more necessary the development of sampling schemes for wideband radio signals with spectral holes. This is equivalent to considering a sparse multiband signal in the framework of Compressive Sampling theory. Starting from previous results on multicoset sampling and recent advances in compressive sampling, we analyze the matrix involved in the corresponding reconstruction equation and define a new method for the design of universal multicoset codes, that is, codes guaranteeing perfect reconstruction of the sparse multiband signal.

Temperature Sensor Placement Including Routing Overhead and Sampling Inaccuracies

Dynamic thermal management techniques require a collection of on-chip thermal sensors that imply a significant area and power overhead. Finding the optimum number of temperature monitors and their location on the chip surface to optimize accuracy is an NP-hard problem. In this work we improve the modeling of the problem by including area, power and networking constraints along with the consideration of three inaccuracy terms: spatial errors, sampling rate errors and monitor-inherent errors. The problem is solved by the simulated annealing algorithm. We apply the algorithm to a test case employing three different types of monitors to highlight the importance of the different metrics. Finally we present a case study of the Alpha 21364 processor under two different constraint scenarios.

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.

Estimation of a probability in inverse binomial sampling under normalized linear-linear and inverse-linear loss

Sequential estimation of the success probability $p$ in inverse binomial sampling is considered in this paper. For any estimator $\hatvap$, its quality is measured by the risk associated with normalized loss functions of linear-linear or inverse-linear form. These functions are possibly asymmetric, with arbitrary slope parameters $a$ and $b$ for $\hatvap < p$ and $\hatvap > p$ respectively. Interest in these functions is motivated by their significance and potential uses, which are briefly discussed. Estimators are given for which the risk has an asymptotic value as $p \rightarrow 0$, and which guarantee that, for any $p \in (0,1)$, the risk is lower than its asymptotic value. This allows selecting the required number of successes, $\nnum$, to meet a prescribed quality irrespective of the unknown $p$. In addition, the proposed estimators are shown to be approximately minimax when $a/b$ does not deviate too much from $1$, and asymptotically minimax as $\nnum \rightarrow \infty$ when $a=b$.

Almost rejectionless sampling from Nakagami-m distributions (m ≥ 1)

The Nakagami-m distribution is widely used for the simulation of fading channels in wireless communications. A novel, simple and extremely efficient acceptance-rejection algorithm is introduced for the generation of independent Nakagami-m random variables. The proposed method uses another Nakagami density with a half-integer value of the fading parameter, mp ¼ n/2 ≤ m, as proposal function, from which samples can be drawn exactly and easily. This novel rejection technique is able to work with arbitrary values of m ≥ 1, average path energy, V, and provides a higher acceptance rate than all currently available methods. RESUMEN. Método extremadamente eficiente para generar variables aleatorias de Nakagami (utilizadas para modelar el desvanecimiento en canales de comunicaciones móviles) basado en "rejection sampling".