Maximum Likelihood Estimation of Modal Parameters in Structures Using the Expectation Maximization Algorithm

This paper presents a time-domain stochastic system identification method based on maximum likelihood estimation (MLE) with the expectation maximization (EM) algorithm. The effectiveness of this structural identification method is evaluated through numerical simulation in the context of the ASCE benchmark problem on structural health monitoring. The benchmark structure is a four-story, two-bay by two-bay steel-frame scale model structure built in the Earthquake Engineering Research Laboratory at the University of British Columbia, Canada. This paper focuses on Phase I of the analytical benchmark studies. A MATLAB-based finite element analysis code obtained from the IASC-ASCE SHM Task Group web site is used to calculate the dynamic response of the prototype structure. A number of 100 simulations have been made using this MATLAB-based finite element analysis code in order to evaluate the proposed identification method. There are several techniques to realize system identification. In this work, stochastic subspace identification (SSI)method has been used for comparison. SSI identification method is a well known method and computes accurate estimates of the modal parameters. The principles of the SSI identification method has been introduced in the paper and next the proposed MLE with EM algorithm has been explained in detail. The advantages of the proposed structural identification method can be summarized as follows: (i) the method is based on maximum likelihood, that implies minimum variance estimates; (ii) EM is a computational simpler estimation procedure than other optimization algorithms; (iii) estimate more parameters than SSI, and these estimates are accurate. On the contrary, the main disadvantages of the method are: (i) EM algorithm is an iterative procedure and it consumes time until convergence is reached; and (ii) this method needs starting values for the parameters. Modal parameters (eigenfrequencies, damping ratios and mode shapes) of the benchmark structure have been estimated using both the SSI method and the proposed MLE + EM method. The numerical results show that the proposed method identifies eigenfrequencies, damping ratios and mode shapes reasonably well even in the presence of 10% measurement noises. These modal parameters are more accurate than the SSI estimated modal parameters.

Computational Schemes for Optimizing Domains of Attraction in Dynamical Systems

In this paper, several computational schemes are presented for the optimal tuning of the global behavior of nonlinear dynamical sys- tems. Specifically, the maximization of the size of domains of attraction associated with invariants in parametrized dynamical sys- tems is addressed. Cell Mapping (CM) tech- niques are used to estimate the size of the domains, and such size is then maximized via different optimization tools. First, a ge- netic algorithm is tested whose performance shows to be good for determining global maxima at the expense of high computa- tional cost. Secondly, an iterative scheme based on a Stochastic Approximation proce- dure (the Kiefer-Wolfowitz algorithm) is eval- uated showing acceptable performance at low cost. Finally, several schemes combining neu- ral network based estimations and optimiza- tion procedures are addressed with promising results. The performance of the methods is illus- trated with two applications: first on the well-known van der Pol equation with stan- dard parametrization, and second the tuning of a controller for saturated systems.

Determinant factors to discriminate the body weight loss in overweight people

Introduction. Most studies have described how the weight loss is when different treatments are compared (1-3), while others have also compared the weight loss by sex (4), or have taken into account psychosocial (5) and lifestyle (6, 7) variables. However, no studies have examined the interaction of different variables and the importance of them in the weight loss. Objective. Create a model to discriminate the range of weight loss, determining the importance of each variable. Methods. 89 overweight people (BMI: 25-29.9 kg?m-2), aged from 18 to 50 years, participated in the study. Four types of treatments were randomly assigned: strength training (S), endurance training (E), strength and endurance training (SE), and control group (C). All participants followed a 25% calorie restriction diet. Two multivariate discriminant models including the variables age, sex, height, daily energy expenditure (EE), type of treatment (T), caloric restriction (CR), initial body weight (BW), initial fat mass (FM), initial muscle mass (MM) and initial bone mineral density (BMD) were performed having into account two groups: the first and fourth quartile of the % of weight loss in the first model; the groups above and below the mean of the % of weight loss in the second model. The discriminant models were built using the inclusion method in SPSS allowing us to find a function that could predict the body weight loss range that an overweight person could achieve in a 6 months weight loss intervention.Results. The first discriminant analysis predicted that a combination of the studied variables would discriminate between the two ranges of body weight loss with 81.4% of correct classification. The discriminant function obtained was (Wilks? Lambda=0.475, p=0.003): Discriminant score=-18.266-(0.060xage)- (1.282xsex[0=female;1=male])+(14.701xheight)+(0.002xEE)- (0.006xT[1=S;2=E;3=SE;4=C])-(0.047xCR)- (0.558xBW)+(0.475xFM)+(0.398xMM)+(3.499xBMD) The second discriminant model obtained would discriminate between the two groups of body weight loss with 74.4% of correct classification. The discriminant function obtained was (Wilks? Lambda=0.725, p=0.005): Discriminant score=-5.021-(0.052xage)- (0.543xsex[0=female;1=male])+(3.530xheight)+(0.001xEE)- (0.493xT[1=S;2=E;3=SE;4=C])+(0.003xCR)- (0.365xBW)+(0.368xFM)+(0.296xMM)+(4.034xBMD) Conclusion. The first developed model could predict the percentage of weight loss in the following way: if the discriminant score is close to 1.051, the range of weight loss will be from 7.44 to -4.64% and if it is close to - 1.003, the range will be from -11.03 to -25,00% of the initial body weight. With the second model if the discriminant score is close to 0.623 the body weight loss will be above -7.93% and if it is close to -0.595 will be below - 7.93% of the initial body weight. References. 1. Brochu M, et al. Resistance training does not contribute to improving the metabolic profile after a 6-month weight loss program in overweight and obese postmenopausal women. J Clin Endocrinol Metab. 2009 Sep;94(9):3226-33. 2. Del Corral P, et al. Effect of dietary adherence with or without exercise on weight loss: a mechanistic approach to a global problem. J Clin Endocrinol Metab. 2009 May;94(5):1602-7. 3. Larson-Meyer DE, et al. Caloric Restriction with or without Exercise: The Fitness vs. Fatness Debate. Med Sci Sports Exerc. 2010;42(1):152-9. 4. Hagan RD, et al. The effects of aerobic conditioning and/or caloric restriction in overweight men and women. Medicine & Science in Sports & Exercise. 1986;18(1):87-94. 5. Teixeira PJ, et al. Mediators of weight loss and weight loss maintenance in middle-aged women. Obesity (Silver Spring). 2010 Apr;18(4):725-35. 6. Bautista-Castano I, et al. Variables predictive of adherence to diet and physical activity recommendations in the treatment of obesity and overweight, in a group of Spanish subjects. Int J Obes Relat Metab Disord. 2004 May;28(5):697-705.

Efficient methodologies for system matrix modelling in iterative image reconstruction for rotating high-resolution PET

A fully 3D iterative image reconstruction algorithm has been developed for high-resolution PET cameras composed of pixelated scintillator crystal arrays and rotating planar detectors, based on the ordered subsets approach. The associated system matrix is precalculated with Monte Carlo methods that incorporate physical effects not included in analytical models, such as positron range effects and interaction of the incident gammas with the scintillator material. Custom Monte Carlo methodologies have been developed and optimized for modelling of system matrices for fast iterative image reconstruction adapted to specific scanner geometries, without redundant calculations. According to the methodology proposed here, only one-eighth of the voxels within two central transaxial slices need to be modelled in detail. The rest of the system matrix elements can be obtained with the aid of axial symmetries and redundancies, as well as in-plane symmetries within transaxial slices. Sparse matrix techniques for the non-zero system matrix elements are employed, allowing for fast execution of the image reconstruction process. This 3D image reconstruction scheme has been compared in terms of image quality to a 2D fast implementation of the OSEM algorithm combined with Fourier rebinning approaches. This work confirms the superiority of fully 3D OSEM in terms of spatial resolution, contrast recovery and noise reduction as compared to conventional 2D approaches based on rebinning schemes. At the same time it demonstrates that fully 3D methodologies can be efficiently applied to the image reconstruction problem for high-resolution rotational PET cameras by applying accurate pre-calculated system models and taking advantage of the system's symmetries.

An approach to operational modal analysis using the expectation maximization algorithm

This paper presents the Expectation Maximization algorithm (EM) applied to operational modal analysis of structures. The EM algorithm is a general-purpose method for maximum likelihood estimation (MLE) that in this work is used to estimate state space models. As it is well known, the MLE enjoys some optimal properties from a statistical point of view, which make it very attractive in practice. However, the EM algorithm has two main drawbacks: its slow convergence and the dependence of the solution on the initial values used. This paper proposes two different strategies to choose initial values for the EM algorithm when used for operational modal analysis: to begin with the parameters estimated by Stochastic Subspace Identification method (SSI) and to start using random points. The effectiveness of the proposed identification method has been evaluated through numerical simulation and measured vibration data in the context of a benchmark problem. Modal parameters (natural frequencies, damping ratios and mode shapes) of the benchmark structure have been estimated using SSI and the EM algorithm. On the whole, the results show that the application of the EM algorithm starting from the solution given by SSI is very useful to identify the vibration modes of a structure, discarding the spurious modes that appear in high order models and discovering other hidden modes. Similar results are obtained using random starting values, although this strategy allows us to analyze the solution of several starting points what overcome the dependence on the initial values used.

Signal to noise ratio maximization in quiet zone acquisitions for range assessment at sub-millimeter wavelengths

This paper proposes a quiet zone probing approach which deals with low dynamic range quiet zone acquisitions. Lack of dynamic range is a feature of millimeter and sub-millimeter wavelength technologies. It is consequence of the gradually smaller power generated by the instrumentation, that follows a f^α law with frequency, being α≥1 variable depending on the signal source’s technology. The proposed approach is based on an optimal data reduction scenario which redounds in a maximum signal to noise ratio increase for the signal pattern, with minimum information losses. After theoretical formulation, practical applications of the technique are proposed.

A comparison between PML, infinite elements and an iterative BEM as mesh truncation methods for HP self-adaptive procedures in electromagnetics

Finite element hp-adaptivity is a technology that allows for very accurate numerical solutions. When applied to open region problems such as radar cross section prediction or antenna analysis, a mesh truncation method needs to be used. This paper compares the following mesh truncation methods in the context of hp-adaptive methods: Infinite Elements, Perfectly Matched Layers and an iterative boundary element based methodology. These methods have been selected because they are exact at the continuous level (a desirable feature required by the extreme accuracy delivered by the hp-adaptive strategy) and they are easy to integrate with the logic of hp-adaptivity. The comparison is mainly based on the number of degrees of freedom needed for each method to achieve a given level of accuracy. Computational times are also included. Two-dimensional examples are used, but the conclusions directly extrapolated to the three dimensional case.

Iterative determination of clinical beam phase space from dose measurements

Monte Carlo (MC) method can accurately compute the dose produced by medical linear accelerators. However, these calculations require a reliable description of the electron and/or photon beams delivering the dose, the phase space (PHSP), which is not usually available. A method to derive a phase space model from reference measurements that does not heavily rely on a detailed model of the accelerator head is presented. The iterative optimization process extracts the characteristics of the particle beams which best explains the reference dose measurements in water and air, given a set of constrains

Lifetime maximization for beamforming applications in wireless sensor networks

Energy efficiency is a major design issue in the context of Wireless Sensor Networks (WSN). If data is to be sent to a far-away base station, collaborative beamforming by the sensors may help to dis- tribute the load among the nodes and reduce fast battery depletion. However, collaborative beamforming techniques are far from opti- mality and in many cases may be wasting more power than required. In this contribution we consider the issue of energy efficiency in beamforming applications. Using a convex optimization framework, we propose the design of a virtual beamformer that maximizes the network's lifetime while satisfying a pre-specified Quality of Service (QoS) requirement. A distributed consensus-based algorithm for the computation of the optimal beamformer is also provided

Operational Modal Analysis using Expectation Maximization Algorithm

This paper presents a time-domain stochastic system identification method based on Maximum Likelihood Estimation and the Expectation Maximization algorithm. The effectiveness of this structural identification method is evaluated through numerical simulation in the context of the ASCE benchmark problem on structural health monitoring. Modal parameters (eigenfrequencies, damping ratios and mode shapes) of the benchmark structure have been estimated applying the proposed identification method to a set of 100 simulated cases. The numerical results show that the proposed method estimates all the modal parameters reasonably well in the presence of 30% measurement noise even. Finally, advantages and disadvantages of the method have been discussed.

El fitness como factor determinante en el riesgo de discapacidad, su relación con la composición corporal y el estado cognitivo en población mayor institucionalizada y de vida independiente = Fitness as a determinant factor in the risk of disability, and its relationship to body composition and cognitive status in institutionalized elderly population and independent living

La condición física, o como mejor se la conoce hoy en día el “fitness”, es una variable que está cobrando gran protagonismo, especialmente desde la perspectiva de la salud. La mejora de la calidad de vida que se ha experimentado en los últimos años en las sociedades desarrolladas, conlleva un aumento de la esperanza de vida, lo que hace que cada vez más personas vivan más años. Este rápido crecimiento de la población mayor de 60 años hace que, un grupo poblacional prácticamente olvidado desde el punto de vista de la investigación científica en el campo de la actividad física y del deporte, cobre gran relevancia, con el fin de poder ayudar a alcanzar el dicho “no se trata de aportar años a la vida sino vida a lo años”. La presente memoria de Tesis Doctoral tiene como principal objetivo valorar los niveles de fitness en población mayor española, además de analizar la relación existente entre el fitness, sus condicionantes y otros aspectos de la salud, tales como la composición corporal y el estado cognitivo. Entendemos que para poder establecer futuras políticas de salud pública en relación a la actividad física y el envejecimiento activo es necesario conocer cuáles son los niveles de partida de la población mayor en España y sus condicionantes. El trabajo está basado en los datos del estudio multicéntrico EXERNET (Estudio Multi-céntrico para la Evaluación de los Niveles de Condición Física y su relación con Estilos de Vida Saludables en población mayor española no institucionalizada), así como en los datos de dos estudios, llevados a cabo en población mayor institucionalizada. Se han analizado un total de 3136 mayores de vida independiente, procedentes de 6 comunidades autónomas, y 153 mayores institucionalizados en residencias de la Comunidad de Madrid. Los principales resultados de esta tesis son los siguientes: a) Fueron establecidos los valores de referencia, así como las curvas de percentiles, para cada uno de los test de fitness, de acuerdo a la edad y al sexo, en población mayor española de vida independiente y no institucionalizada. b) Los varones obtuvieron mejores niveles de fitness que las mujeres, excepto en los test de flexibilidad; existe una tendencia a disminuir la condición física en ambos sexos a medida que la edad aumenta. c) Niveles bajos de fitness funcional fueron asociados con un aumento en la percepción de problemas. d) El nivel mínimo de fitness funcional a partir del cual los mayores perciben problemas en sus actividades de la vida diaria (AVD) es similar en ambos sexos. e) Niveles elevados de fitness fueron asociados con un menor riesgo de sufrir obesidad sarcopénica y con una mejor salud percibida en los mayores. f) Las personas mayores con obesidad sarcopénica tienen menor capacidad funcional que las personas mayores sanas. g) Niveles elevados de fuerza fueron asociados con un mejor estado cognitivo siendo el estado cognitivo la variable que más influye en el deterioro de la fuerza, incluso más que el sexo y la edad. ABSTRACT Fitness is a variable that is gaining in prominence, especially from the health perspective. Improvement of life quality that has been experienced in the last few years in developed countries, leads to an expanded life expectancy, increasing the numbers of people living longer. This population consisting of people of over 60 years, an almost forgotten population group from the point of view of scientific research in the field of physical activity and sport, is becoming increasingly important, with the main aim of helping to achieve the saying “do not only add years to life, but also add life to years”. The principal aim of the current thesis was to assess physical fitness levels in Spanish elderly people, of over 65 years, analyzing relationship between physical fitness, its determinants, and other aspects of health such as body composition and cognitive status. In order to establish further public health policies in relation to physical activity and active ageing it is necessary to identify the starting physical fitness levels of the Spanish population and their determinants. The work is based on data from the EXERNET multi-center study ("Multi-center Study for the Evaluation of Fitness levels and their relationship to Healthy Lifestyles in noninstitutionalized Spanish elderly"), and on data from two studies conducted in institutionalized elderly people: a total of 3136 non-institutionalized elderly, from 6 Regions of Spain, and 153 institutionalized elderly in nursing homes of Madrid. The main outcomes of this thesis are: a) sex- and age-specific physical fitness normative values and percentile curves for independent and non-institutionalized Spanish elderly were established. b) Greater physical fitness was present in the elderly men than in women, except for the flexibility test, and a trend toward decreased physical fitness in both sexes as their age increased. c) Lower levels of functional fitness were associated with increased perceived problems. d) The minimum functional fitness level at which older adults perceive problems in their ADLs, is similar for both sexes e) Higher levels of physical fitness were associated with a reduced risk of suffering sarcopenic obesity and better perceived health among the elderly. f) The elderly with sarcopenic obesity have lower physical functioning than healthy counterparts. g) Higher strength values were associated with better cognitive status with cognitive status being the most influencing variable in strength deterioration even more than sex and age.

Fuel poverty as a determinant in energy retrofitting actions

Fuel poverty can be defined as “the inability to afford adequate warmth in the home" and it is the result of the combination of three items: low household income, housing lack of energy efficiency and high energy bills. Although it affects a growing number of households within the European Union only some countries have an official definition for it. In 2013, the European Parliament claimed the Commission and Estate Members to develop different policies in order to fight household energy vulnerability. The importance of tackling fuel poverty is based on the critical consequences it has for human health living below certain temperatures. In Spain some advances have been made in this field but main existing studies remain at the statistical level and do not deepen the understanding of the problem from the perspective of dwelling indoor habitability conditions. What is more, this concept is yet to be officially defined. This paper presents the evaluation of fuel poverty in a building block of social housing located in the centre of Zaragoza and how this issue determined the strategies implemented in the energy retrofitting intervention project. At a first step, fuel poverty was appraised through the exploration of indoor thermal conditions. The adaptive thermal comfort (UNE-EN 15251:2008) method was used to establish the appropriate indoor temperatures and consequently to determine what can be called 'comfort gap'. Results were collated and verified with energy bills collection and a survey work that gathered data from neighbours. All this permitted pointing out those households more in need. Results from the social analysis combined with the evaluation of the building thermal performance determined the intervention. The renovation project was aimed at the implementation of passive strategies that improve households thermal comfort in order to alleviate households fuel poverty situation. This research is part of the project NewSolutions4OldHousing (LIFE10 ENV/ES/439) cofounded by the European Commission under the LIFE+ Programme.

Iterative bit- and power allocation in correlated MIMO systems

In this contribution a novel iterative bit- and power allocation (IBPA) approach has been developed when transmitting a given bit/s/Hz data rate over a correlated frequency non-selective (4× 4) Multiple-Input MultipleOutput (MIMO) channel. The iterative resources allocation algorithm developed in this investigation is aimed at the achievement of the minimum bit-error rate (BER) in a correlated MIMO communication system. In order to achieve this goal, the available bits are iteratively allocated in the MIMO active layers which present the minimum transmit power requirement per time slot.