Adaptive Markov chain Monte Carlo and Bayesian filtering for state space models

This thesis is concerned with the state and parameter estimation in state space models. The estimation of states and parameters is an important task when mathematical modeling is applied to many different application areas such as the global positioning systems, target tracking, navigation, brain imaging, spread of infectious diseases, biological processes, telecommunications, audio signal processing, stochastic optimal control, machine learning, and physical systems. In Bayesian settings, the estimation of states or parameters amounts to computation of the posterior probability density function. Except for a very restricted number of models, it is impossible to compute this density function in a closed form. Hence, we need approximation methods. A state estimation problem involves estimating the states (latent variables) that are not directly observed in the output of the system. In this thesis, we use the Kalman filter, extended Kalman filter, Gauss–Hermite filters, and particle filters to estimate the states based on available measurements. Among these filters, particle filters are numerical methods for approximating the filtering distributions of non-linear non-Gaussian state space models via Monte Carlo. The performance of a particle filter heavily depends on the chosen importance distribution. For instance, inappropriate choice of the importance distribution can lead to the failure of convergence of the particle filter algorithm. In this thesis, we analyze the theoretical Lᵖ particle filter convergence with general importance distributions, where p ≥2 is an integer. A parameter estimation problem is considered with inferring the model parameters from measurements. For high-dimensional complex models, estimation of parameters can be done by Markov chain Monte Carlo (MCMC) methods. In its operation, the MCMC method requires the unnormalized posterior distribution of the parameters and a proposal distribution. In this thesis, we show how the posterior density function of the parameters of a state space model can be computed by filtering based methods, where the states are integrated out. This type of computation is then applied to estimate parameters of stochastic differential equations. Furthermore, we compute the partial derivatives of the log-posterior density function and use the hybrid Monte Carlo and scaled conjugate gradient methods to infer the parameters of stochastic differential equations. The computational efficiency of MCMC methods is highly depend on the chosen proposal distribution. A commonly used proposal distribution is Gaussian. In this kind of proposal, the covariance matrix must be well tuned. To tune it, adaptive MCMC methods can be used. In this thesis, we propose a new way of updating the covariance matrix using the variational Bayesian adaptive Kalman filter algorithm.

The radio frequency identification technology in the supply chain management: a case of a big logistics company

Nowadays global business trends force the adoption of innovative ICTs into the supply chain management (SCM). Particularly, the RFID technology is on high demand among SCM professionals due to its business advantages such as improving of accuracy and veloc-ity of SCM processes which lead to decrease of operational costs. Nevertheless, a question of the RFID technology impact on the efficiency of warehouse processes in the SCM re-mains open. The goal of the present study is to experiment the possibility of improvement order picking velocity in a warehouse of a big logistics company with the use of the RFID technology. In order to achieve this goal the following objectives have been developed: 1) Defining the scope of the RFID technology applications in the SCM; 2) Justification of the RFID technology impact on the SCM processes; 3) Defining a place of the warehouse order picking process in the SCM; 4) Identification and systematization of existing meth-ods of order picking velocity improvement; 5) Choosing of the study object and gathering of the empirical data about number of orders, number of hours spent per each order line daily during 5 months; 6) Processing and analysis of the empirical data; 7) Conclusion about the impact of the RFID technology on the speed of order picking process. As a result of the research it has been found that the speed of the order picking processes has not been changed as time has gone after the RFID adoption. It has been concluded that in order to achieve a positive effect in the speed of order picking process with the use of the RFID technology it is necessary to simultaneously implement changes in logistics and organizational management in 3PL logistics companies. Practical recommendations have been forwarded to the management of the company for further investigation and procedure.

Radio-frequency-based measurement methods for bearing current analysis in induction motors

The most common reason for a low-voltage induction motor breakdown is a bearing failure. Along with the increasing popularity of modern frequency converters, bearing failures have become the most important motor fault type. Conditions in which bearing currents are likely to occur are generated as a side effect of fast du/dt switching transients. Once present, different types of bearing currents can accelerate the mechanical wear of bearings by causing deformation of metal parts in the bearing and degradation of the lubricating oil properties.The bearing current phenomena are well known, and several bearing current measurement and mitigation methods have been proposed. Nevertheless, in order to develop more feasible methods to measure and mitigate bearing currents, better knowledge of the phenomena is required. When mechanical wear is caused by bearing currents, the resulting aging impact has to be monitored and dealt with. Moreover, because of the stepwise aging mechanism, periodically executed condition monitoring measurements have been found ineffective. Thus, there is a need for feasible bearing current measurement methods that can be applied in parallel with the normal operation of series production drive systems. In order to reach the objectives of feasibility and applicability, nonintrusive measurement methods are preferred. In this doctoral dissertation, the characteristics and conditions of bearings that are related to the occurrence of different kinds of bearing currents are studied. Further, the study introduces some nonintrusive radio-frequency-signal-based approaches to detect and measure parameters that are associated with the accelerated bearing wear caused by bearing currents.

Study of Control Methods of a High-Frequency Cycloconverter for Domestic Induction Heating Appliances

A high-frequency cyclonverter acts as a direct ac-to-ac power converter circuit that does not require a diode bidge rectifier. Bridgeless topology makes it possible to remove forward voltage drop losses that are present in a diode bridge. In addition, the on-state losses can be reduced to 1.5 times the on-state resistance of switches in half-bridge operation of the cycloconverter. A high-frequency cycloconverter is reviewed and the charging effect of the dc-capacitors in ``back-to-back'' or synchronous mode operation operation is analyzed. In addition, a control method is introduced for regulating dc-voltage of the ac-side capacitors in synchronous operation mode. The controller regulates the dc-capacitors and prevents switches from reaching overvoltage level. This can be accomplished by variating phase-shift between the upper and the lower gate signals. By adding phase-shift between the gate signal pairs, the charge stored in the energy storage capacitors can be discharged through the resonant load and substantially, the output resonant current amplitude can be improved. The above goals are analyzed and illustrated with simulation. Theory is supported with practical measurements where the proposed control method is implemented in an FPGA device and tested with a high-frequency cycloconverter using super-junction power MOSFETs as switching devices.

Low frequency of p53 mutations in cervical carcinomas among Brazilian women

Human papillomavirus (HPV) infections of the high-risk types are strongly linked to the development of cervical carcinoma. The HPV oncoproteins E6 and E7 are thought to play a crucial role in this process through their interactions with the p53 protein and the retinoblastoma susceptibility gene product pRb, respectively. E6 binds to p53 protein promoting its degradation. This is considered to contribute to the oncogenesis of HPV-associated anogenital cancer. On the other hand, in HPV-negative cervical carcinoma, p53 mutations are thought to have a role in the transformation process. A total of 122 HPV-positive cervical carcinoma tissue samples were evaluated for the presence of mutations in exons 5-8 of the p53 gene by single-stranded conformation polymorphism analysis and DNA sequencing. Only four missense point mutations were detected. These findings suggest that other mechanisms independent of p53 inactivation may play a role in the genesis of cervical carcinomas.

Dietary fiber intake, stool frequency and colonic transit time in chronic functional constipation in children

The objective of the present study was to evaluate associations between fiber intake, colonic transit time and stool frequency. Thirty-eight patients aged 4 to 14 years were submitted to alimentary evaluation and to measurement of colonic transit time. The median fiber intake of the total sample was age + 10.3 g/day. Only 18.4% of the subjects presented a daily dietary fiber intake below the levels recommended by the American Health Foundation. In this group, the median left colonic transit time was shorter than in the group with higher dietary fiber intake (11 vs 17 h, P = 0.067). The correlation between stool frequency and colonic transit time was negative and weak for left colon (r = -0.3, P = 0.04), and negative and moderate for rectosigmoid and total colon (r = -0.5, P<0.001 and r = -0.5, P<0.001, respectively). The stool frequency was lower in the group with slow transit time (0.8 vs 2.3 per week, P = 0.014). In conclusion, most patients with chronic functional constipation had adequate dietary fiber intake. The negative correlation between stool frequency and colonic transit time increased progressively from proximal segments to distal segments of the colon. Patients with normal and prolonged colonic transit time differ in terms of stool frequency.

Narrow-band 1, 2, 3, 4, 8, 16 and 24 cycles/360º angular frequency filters

We measured human frequency response functions for seven angular frequency filters whose test frequencies were centered at 1, 2, 3, 4, 8, 16 or 24 cycles/360º using a supra-threshold summation method. The seven functions of 17 experimental conditions each were measured nine times for five observers. For the arbitrarily selected filter phases, the maximum summation effect occurred at test frequency for filters at 1, 2, 3, 4 and 8 cycles/360º. For both 16 and 24 cycles/360º test frequencies, maximum summation occurred at the lower harmonics. These results allow us to conclude that there are narrow-band angular frequency filters operating somehow in the human visual system either through summation or inhibition of specific frequency ranges. Furthermore, as a general result, it appears that addition of higher angular frequencies to lower ones disturbs low angular frequency perception (i.e., 1, 2, 3 and 4 cycles/360º), whereas addition of lower harmonics to higher ones seems to improve detection of high angular frequency harmonics (i.e., 8, 16 and 24 cycles/360º). Finally, we discuss the possible involvement of coupled radial and angular frequency filters in face perception using an example where narrow-band low angular frequency filters could have a major role.

High frequency of the CCR5delta32 variant among individuals from an admixed Brazilian population with sickle cell anemia

Homozygous sickle cell disease (SCD) has a wide spectrum of clinical manifestations. In Brazil, the main cause of death of individuals with SCD is recurrent infection. The CCR5delta32 allele, which confers relative resistance to macrophage-tropic HIV virus infection, probably has reached its frequency and world distribution due to other pathogens that target macrophage in European populations. In the present investigation a relatively higher prevalence (5.1%) of the CCR5delta32 allele was identified, by PCR amplification using specific primers, in 79 SCD patients when compared to healthy controls (1.3%) with the same ethnic background (Afro-Brazilians). Based on a hypothesis that considers SCD as a chronic inflammatory condition, and since the CCR5 chemokine receptor is involved in directing a Th1-type immune response, we suggest that a Th1/Th2 balance can influence the morbidity of SCD. If the presence of the null CCR5delta32 allele results in a reduction of the chronic inflammation state present in SCD patients, this could lead to differential survival of SCD individuals who are carriers of the CCR5delta32 allele. This differential survival could be due to the development of less severe infections and consequently reduced or less severe vaso-occlusive crises.

Effect of lung resection and sham surgery on the frequency of infection in alloxan-diabetic rats

The present study was carried out in order to determine the effect of lung resection on the frequency of infections in alloxan-diabetic rats. Adult female Wistar rats were injected with alloxan (40 mg/kg, iv) to induce diabetes mellitus (group D; N = 45) or with vehicle (1.0 ml/kg, iv) to be used as controls (group C; N = 45). Thirty-six days after receiving alloxan both groups were randomly divided into three subgroups: no operation (NO; N = 15), sham operation (SO; N = 15), and left pneumonectomy (PE; N = 15). The rats were sacrificed 36 days after surgery and their lungs were examined microscopically and macroscopically. The occurrence of thoracic wall infection, thoracic wall abscess, lung abscess and pleural empyema was similar in groups D and C. In contrast, the overall infection rate was higher (P<0.05) in the diabetic rats (SO-D and PE-D subgroups, but not in the NO-D subgroup). Considering that the overall infection rate was similar in the SO-D and PE-D subgroups, we suggest that surgery but not pneumonectomy was related to the higher prevalence of infection in diabetic rats.

Virtual Runtime Application Partitions for Resource Management in Massively Parallel Architectures

This thesis presents a novel design paradigm, called Virtual Runtime Application Partitions (VRAP), to judiciously utilize the on-chip resources. As the dark silicon era approaches, where the power considerations will allow only a fraction chip to be powered on, judicious resource management will become a key consideration in future designs. Most of the works on resource management treat only the physical components (i.e. computation, communication, and memory blocks) as resources and manipulate the component to application mapping to optimize various parameters (e.g. energy efficiency). To further enhance the optimization potential, in addition to the physical resources we propose to manipulate abstract resources (i.e. voltage/frequency operating point, the fault-tolerance strength, the degree of parallelism, and the configuration architecture). The proposed framework (i.e. VRAP) encapsulates methods, algorithms, and hardware blocks to provide each application with the abstract resources tailored to its needs. To test the efficacy of this concept, we have developed three distinct self adaptive environments: (i) Private Operating Environment (POE), (ii) Private Reliability Environment (PRE), and (iii) Private Configuration Environment (PCE) that collectively ensure that each application meets its deadlines using minimal platform resources. In this work several novel architectural enhancements, algorithms and policies are presented to realize the virtual runtime application partitions efficiently. Considering the future design trends, we have chosen Coarse Grained Reconfigurable Architectures (CGRAs) and Network on Chips (NoCs) to test the feasibility of our approach. Specifically, we have chosen Dynamically Reconfigurable Resource Array (DRRA) and McNoC as the representative CGRA and NoC platforms. The proposed techniques are compared and evaluated using a variety of quantitative experiments. Synthesis and simulation results demonstrate VRAP significantly enhances the energy and power efficiency compared to state of the art.

Adaptive Dynamics of Resource Specialization

Ecological specialization in resource utilization has various facades ranging from nutritional resources via host use of parasites or phytophagous insects to local adaptation in different habitats. Therefore, the evolution of specialization affects the evolution of most other traits, which makes it one of the core issues in the theory of evolution. Hence, the evolution of specialization has gained enormous amounts of research interest, starting already from Darwin’s Origin of species in 1859. Vast majority of the theoretical studies has, however, focused on the mathematically most simple case with well-mixed populations and equilibrium dynamics. This thesis explores the possibilities to extend the evolutionary analysis of resource usage to spatially heterogeneous metapopulation models and to models with non-equilibrium dynamics. These extensions are enabled by the recent advances in the field of adaptive dynamics, which allows for a mechanistic derivation of the invasion-fitness function based on the ecological dynamics. In the evolutionary analyses, special focus is set to the case with two substitutable renewable resources. In this case, the most striking questions are, whether a generalist species is able to coexist with the two specialist species, and can such trimorphic coexistence be attained through natural selection starting from a monomorphic population. This is shown possible both due to spatial heterogeneity and due to non-equilibrium dynamics. In addition, it is shown that chaotic dynamics may sometimes inflict evolutionary suicide or cyclic evolutionary dynamics. Moreover, the relations between various ecological parameters and evolutionary dynamics are investigated. Especially, the relation between specialization and dispersal propensity turns out to be counter-intuitively non-monotonous. This observation served as inspiration to the analysis of joint evolution of dispersal and specialization, which may provide the most natural explanation to the observed coexistence of specialist and generalist species.

Contrast sensitivity to angular frequency gratings is not higher than to Cartesian gratings

When contrast sensitivity functions to Cartesian and angular gratings were compared in previous studies the peak sensitivity to angular stimuli was reported to be 0.21 log units higher. In experiments carried out to repeat this result, we used the same two-alternative forced-choice paradigm, but improved experimental control and precision by increasing contrast resolution from 8 to 12 bits, increasing the screen refresh rate from 30 Hz interlaced to 85 Hz non-interlaced, linearizing the voltage-luminance relation, modulating luminance in frequencies that minimize pixel aliasing, and improving control of the subject's exposure to the stimuli. The contrast sensitivity functions to Cartesian and angular gratings were similar in form and peak sensitivity (2.4 cycles per visual degree (c/deg) and 32 c/360º, respectively) to those reported in a previous study (3 c/deg and 32 c/360º, respectively), but peak sensitivity to angular stimuli was 0.13 log units lower than that to Cartesian stimuli. When the experiment was repeated, this time simulating the experimental control level used in the previous study, no difference between the peak sensitivity to Cartesian and angular stimuli was found. This result agrees with most current models that assume Cartesian filtering at the first visual processing stage. The discrepancy in the results is explained in part by differences in the degree of experimental control.

Comparison of angular frequency contrast sensitivity in young and older adults

The aim of the present study was to measure contrast sensitivity curves for angular frequencies in the range between 2 and 96 cycles/360º in older human adult volunteers and to compare these measurements with the more usual contrast sensitivity functions for sine-wave gratings. All subjects were free of identifiable ocular disease and had normal acuity. We measured the contrast thresholds for young adults (N = 6; age range, 20-26 years) and older adults (N = 6; age range, 60-67 years) using the psychophysical forced-choice method. In this paradigm the volunteers had to choose the stimulus containing a test frequency at low contrast (e.g., either a sine-wave grating or an angular frequency stimulus), or another neutral stimulus at mean luminance (without any contrast). Older adults presented a loss in contrast sensitivity at high and medium angular frequencies compared to the young adults (i.e., from 8 to 96 cycles/360º). Contrary to expectation, contrast sensitivity at low angular frequencies, i.e., 2 and 4 cycles/360º, was better for the older group than for the younger group. On the other hand, contrast sensitivity for sine-wave gratings at 3 and 4 cpd was higher for young adults as expected. These results suggest age-related changes in the contrast sensitivity function for angular frequencies.

Radial frequency stimuli and sine-wave gratings seem to be processed by distinct contrast brain mechanisms

An assumption commonly made in the study of visual perception is that the lower the contrast threshold for a given stimulus, the more sensitive and selective will be the mechanism that processes it. On the basis of this consideration, we investigated contrast thresholds for two classes of stimuli: sine-wave gratings and radial frequency stimuli (i.e., j0 targets or stimuli modulated by spherical Bessel functions). Employing a suprathreshold summation method, we measured the selectivity of spatial and radial frequency filters using either sine-wave gratings or j0 target contrast profiles at either 1 or 4 cycles per degree of visual angle (cpd), as the test frequencies. Thus, in a forced-choice trial, observers chose between a background spatial (or radial) frequency alone and the given background stimulus plus the test frequency (1 or 4 cpd sine-wave grating or radial frequency). Contrary to our expectations, the results showed elevated thresholds (i.e., inhibition) for sine-wave gratings and decreased thresholds (i.e., summation) for radial frequencies when background and test frequencies were identical. This was true for both 1- and 4-cpd test frequencies. This finding suggests that sine-wave gratings and radial frequency stimuli are processed by different quasi-linear systems, one working at low luminance and contrast level (sine-wave gratings) and the other at high luminance and contrast levels (radial frequency stimuli). We think that this interpretation is consistent with distinct foveal only and foveal-parafoveal mechanisms involving striate and/or other higher visual areas (i.e., V2 and V4).

Frequency of Fanconi anemia in Brazil and efficacy of screening for the FANCA 3788-3790del mutation

Fanconi anemia (FA) is an autosomal recessive genetic disease characterized by progressive bone marrow failure, susceptibility to cancer and multiple congenital anomalies. There is important clinical variability among patients and the knowledge of factors which might predict outcome would greatly help the decision making regarding the choices of treatment and the appropriate time to start it. Future studies of the possible correlation between specific mutations with specific clinical presentations will provide the answer to one of these factors. At our Center we standardized a rapid and precise screening test using a mismatch PCR assay for a specific mutation (3788-3790del in exon 38 of gene FANCA) in Brazilian FA patients. We present the results obtained after screening 80 non-consanguineous FA patients referred from all regions of Brazil with a clinical diagnosis of FA supported by cellular hypersensitivity to diepoxybutane. We were able to detect the 3788-3790del allele in 24 of the 80 (30%) FA patients studied. Thirteen of the 80 (16.25%) were homozygotes and 11 of the 80 (13.75%) were compound heterozygotes, thus confirming the high frequency of the FANCA 3788-3790del mutation in Brazilian FA patients. The identification of patients with specific mutations in the FA genes may lead to a better clinical description of this condition, also providing data for genotype-phenotype correlations, to a better understanding of the interaction of this specific mutation with other mutations in compound heterozygote patients, and ultimately to the right choices of treatment for each patient with improvement of the prognosis on future studies.