ELECTRONIC TECHNOLOGY

Author: CARRILLO CALLEJA JUAN MANUEL.
Year: 2001.
University: EXTREMADURA [www.unex.es].
Place of defense: ESCUELA DE INGENIERÍAS INDUSTRIALES.
Place of preparation: FACULTAD DE CIENCIAS ESCUELA DE ING.INDUSTRIALES.

Summary: Advances in CMOS technologies are driving voltages feeding delos integrated circuits values increasingly reducidos.A extent that the dimensions of the devices falling voltages applied need to be scaled in proportion with the aim of ensuring their reliability and controlling the power density dissipated by ellos.Por other hand, from the viewpoint of circuitso analog, many aspects of design that only a few years ago were barely taken into account, paid particular importance in these operating conditions imposed the restricted supply voltages. The operational amplifier is the largest bloc within the analog systems and so mixto.Dichos systems implement a large number of diverse functions, including filtering highlights as one of the most comunes.Este work focuses on design amplifiers rail-to-rail low voltage supply with constant behavior on the entire range of voltaje.Las specific inputs that meets this work include (1). new techniques transconductancia (gm) that are simultaneously robust constant and universal, and therefore, consistent withthe devices CMOS submicrométricos, (2) design techniques to keep constant the behavior of large signal amplifier, ie slew-rate, (3) development stages input operational amplifiers gm constant, constant and SR high bandwidth, and (4) implementing some of the techniques proposed in the design of operational amplifiers capable of operating voltages feeding extremely bajos.Con so validadr benefits experimental techniques all proposals are designed, manufactured and measured different prototypes implementdos in silicon.

Author: VALVERDE VALVERDE JOSE M..
Year: 2002.
University: EXTREMADURA [www.unex.es].
Place of defense: ESCUELA DE INGENIERÍA INDUSTRIALES.
Place of preparation: FACULTAD DE CIENCIAS/ESCUELA DE ING. INDUSTRIALES.

Summary: This work falls within the field of innovation and technological development. Specifically, the contents thereof are the result of collaborations R & D & TD that come with keeping industries in the Extremadura region, and in this regard, we felt it appropriate to focus this work on two sectors whose importance to our regional context it is beyond any dudda. These are the cork industry and ornamental rocks. In relation to the cork industry, there are currently several bottlenecks to those who have not yet provided them with an appropriate solution. But if those overlooking nature químia and / or biological, and we focused on the field of Technologies Avanazadas of Production and Information Technology and Communications, the challenges of the sector are at the two ends of the production chain. At the beginning of the same, it appears the problem of estimation of the characteristics of the cork harvest in the bush alcornocal, and in the end, quality control of its main product derived: the stopper cork. Both problems are provided with respective solutions in this work, which, while not a definitive way, a significant contribution to improving the state of the art. For the first case, an electronic unit that provides accurate information on the most crucial variable in the final application cork bred, as is his caliber. In relation to the second activity, which involves an extraordinary challenge because of the intrinsic heterogeneity of the material, applied for the first time a method of ranking based on processing neuromórfico of information that greatly improves the percentage of hits of traditional systems. Finally, and regarding the sector ornamental rocks, the work focuses on quality control and optimization of a production of roofing slate. In particular, it has developed a system based on image processing which improves, also in a significant way, the current state of the art in the industry, characterized by the use, yet today, exclusively procedures manuals.

Author: VILLARROYA GAUDÓ MARÍA.
Year: 2004.
University: AUTÓNOMA DE BARCELONA [www.uab.es].
Place of defense: ESCOLA TÉCNICA SUPERIOR D'ENGINYERIA.
Place of preparation: ESCUELA DE POSTGRADO.

Summary: The objective of this thesis is the implementation of high-resolution sensors, systems formed by micro / nano electromechanically integrated monolíticamente based on leverage as a transducer, and using technology for the manufacture of silicon. Specifically, it will determine the optimal manufacturing technology for the implementation of sensors based on levers, for applications in vacuum and air or fluid. It will establish the excitation and detection techniques suitable for sensors based on levers. And is there a reconciliation of manufacturing technology of sensors based on levers. And is there a reconciliation of manufacturing technology sensors with CMOS technology, so as to ensure the integration of monolithic system. To that end, produced three different demonstrators, including two gentlemen mass formed by resonant levers and a third system capable of working in liquid medium for electrochemical detection. In the first demonstrator manufactured a sensor mass formed by an array of levers polysilicon integrated monolíticamente with circuitry reading. This layer is used as a structural levels of polysilicon CMOS technology used (technology CMOS CNM25 2P, 2M with two levels of metal and two levels of polysilicon). We have designed arrays of four eight levers that allow measurements mutliplexadas each of the levers independently and differential measures. So on the one hand increases the versatility of the system and to perform differential measures improves the resolution. During the CMOS process defines the areas of manufacturing and post as defined mechanical transducers. After characterization of electrical systems, this demonstrator concludes monolithic integration is possible for a versatile, with resolution mass of less than 40 g / Hz. The second demonstrator consists of a sensor mass formed by resonant levers of crystalline silicon. To use crystalline silicon layer as structural develops a new technology, from SOI substrates, which allows you to define regions for the manufacture of CMOS circuitry and regions with SOI structures for the implementation of transdutores. Once defined technology, sensors are implemented resonant mass (as in the first demonstrator) with the best features of the structural layer. It has tested the operation of these sensors with a high resolution mass of 7 g / Hz. The technology developed allows production systems MEMS / NEMS integrated monolíticamente using crystalline silicon as a structural layer. Lastly has developed a third device, which allows you to work in liquid medium. It is used as a lever transducer element silicon crystal. To detect the deflection of the lever (caused by stress due to surface deposit of móleculas) are measured variations in electrochemical flow between the lever and an electrode is very close to it within a biopotenciostado. There is a need for the separation between two electrodes is less than 100 nm to measure suit. Defining these separations represents a major technological challenge, as it is to define cuts in silicon of a micron thick, with a width of less than 100 nm. They are used lithography techniques with nanometer resolution (microsocopio of atomic forces, AFM and localized beam of ions, IFF) combined with dry recorded by reactive ion (RIE) and direct attack by IFF. There have been cuts required and demonstrated the operation of the device.

Author: COSLADO ARISTIZABAL FRANCISCO JOSÉ.
Year: 2004.
University: MÁLAGA [www.uma.es].
Place of defense: E.T.S.I TELECOMUNICACIÓN.
Place of preparation: ETSI TELECOMUNICACIÓN.

Summary: Compared to the conventional vision systems, vision Foveal is characterized by the collection and management of images multirresolución where the maximum resolution is at the center of the field of vision and is decreasing as we radially away from the center. This technique makes it possible to work with a wide field of vision and high resolution in one or more areas of the image, driving a small data set and thus facilitating the real-time processing of the same. Therefore, their applications will be tasks such as autonomous navigation, recognition and tracking objects in dynamic systems. In return, precise control mechanisms to manage the area to the highest resolution (s) area (s) of interest. Within the field of vision foveal, three-dimensional hierarchical data structures, such as pyramid and the polygon, allowing withstand processing such images multirresolución. This thesis presents the development of a hardware algorithm segmentation of images foveales in a hierarchical structure, the estate, so they are allowed to extract classes or segments of an image in real time (increased 20 images / s). The outcome of the previous process used to perform tasks such as object detection and recognition. The development hardware performed would have the advantage of obtaining a permit processing subsystem that integrates into one card along with the other modules on which it depends, can perform processing medium and low level, freeing the other blocs system components vision foveal for the completion of these tasks, with little occupation of the area, reduced consumption, and speeds hardly achievable through other alternatives.

Author: GARCIA VIZCAINO DAVID.
Year: 2004.
University: POLITÉCNICA DE CATALUÑA [www.upc.edu].
Place of defense: tsc.
Place of preparation: EDIFICI D4 Campus NORD.

Summary: The practical use of the Doppler effect in the laser emission was proposed from the beginning of the development of laser technology in the sixties. Only in the eighties research done could leave the lab and lead to the manufacture of apparatus for measuring speed commercial. Since the nineties these devices is rapidly popularized. Currently used for measuring speed laser Doppler effect in multiple applications, among which stands out the extent of fluid velocity, air and hydrodynamic studies. Its unique features, such as precision obtained so far, the high spatial resolution and non-intrusive, have only recently begun to take account of rivals such as the velocimetría image particles (IVP). Also far speeds of solid beginning to be mobile, with lower overall components opto-electrónicos a target for many companies. Among the applications of this type contemplated control speed of vehicles on the road and control of industrial processes of textile industry, paper manufacturers and cable companies, among others. Companies in Europe and America, as Dantec Electronik and TSI, to cite the most representative, sell equipment LDA general purpose high performance. To date, these systems could only be purchased by major research centers and large companies, due to their high cost. The future of the commercial velocimetría laser Doppler requires the manufacture of cheaper appliances and tailored to customer needs. Many of the current systems are bulky, difficult to handle and powers of work high. It is carrying out an important work to reduce the size and cost of computers without losing their main benefits. Also, the high speed and capacity of calculating personal computers today must make unnecessary the inclusion of specific processors for these teams. We present the design and construction of a comprehensive system for measuring the speed of two components, system 2D-LDA for industrial applications and hydrodynamic low power. Following the philosophy outlined above, the design of our system LDA was performed using only one laser source and a detector module. The LDA systems measure velocity of two components are commercially available, in contrast, employ two-wavelength optical photodetectors and two independents. Emissions typical blue and green lasers ion-Ar are wavelengths often chosen in such measures. Moreover, it has used the two input channels of a card acquisition of general use to perform multilevel shooting. This configuration allowed to work at any given time with the signal burst theoretically better, higher value of signal to noise.

Author: SÁNCHEZ MELENDEZ CESAR.
Year: 2005.
University: POLITÉCNICA DE VALENCIA [www.upv.es].
Place of defense: INGENIERÍA ELECTRÓNICA.
Place of preparation: UNIVERSIDAD POLITÉCNICA DE VALENCIA.

Summary: The atrial fibrillation (AF) is one of the most common arrhythmias and more revenue for emergencies in hospitals currently produces. Today, the processing and analysis of the FA is not entirely satisfactory as that associated with high mortality, reduced quality of life and high health costs, has led to a large increase in publications on this issue in recent years. An efficient non-invasive study of this type of arrhythmia requires the isolated observation of the activity associated headset, usually smaller scale and with covert activity ventiruclar, which precludes the use of linear filtering techniques. Techniques pair extraction activity newest headset, Blind Source Separation, Termination Espacio-Temporal and Cancel Networking Artificial Neurons have a better performance than conventional techniques, theft of Complex East and Recurrent adaptive filtering, but they need the use of the information contained in various derivations, or records with a given duration for best results. The wavelet transform offers a range of possibilities that have not yet been fully explored by the various classic methods of extraction activity headset. So far the main fields of use of wavelets, have been understanding ECG, pattern recognition and elimination of abnormal noise and interference. Therefore, the removal of atrial activity in a non-invasive, wavelet transform is a novel technique. The hypothesis of this dissertation is to verify the potential of the wavelet analysis and the combination of wavelet and blind source separation (BSS), in all its variants, for the removal of the activity headset with electrocardiogram signal from a small number derivations of available, as an alternative to the techniques

Author: RAMOS LARA RAFAEL.
Year: 2005.
University: POLITÉCNICA DE CATALUÑA [www.upc.edu].
Place of defense: AULA DE TELEENSENYAMENT, EDIFICI B3, CAMPUS NORD.
Place of preparation: C4 Jordi Girona, 1-3, campus nord 08034.

Summary: The increase in demand for power supply systems has increased interest in modular power systems operate in parallel, with the first application of the same systems supply uninterrupted. This modular design based on the connected in parallel switched converters has advantages with regard to increasing the power available to the exit, ease of maintenance, increased reliability, capacity expansion, and so on. However, such systems poses difficulty as the design of control circuit allowing a good performance in terms of stability, output ripple, transient response in situations conexión-desconexión modules, and so on. The complexity of the problem of design control increases when one considers that the supply system can be comprised of modules of various features and benefits they receive energy from different sources. According to the explanatory statement earlier in this thesis deals with the problem of designing and conducting a modular system based on AC power connected in parallel modules investors reducers controlled by the technique of sliding mode control to regulate the output voltage AC and equalize the flow through modules investors. In this regard, the methodology has been applied to design control systems for sliding mode multivariables to investors and modular systems have been proposed switching functions that implement strategies equalization flows Master-Slave, Circular Chain Control and the Central Limit Control. Moreover, it has been dealt with the problem of how to control a slide fixed frequency introducing a new algorithm based on the average of zero dynamics (ZAD). The algorithm is obtained under the assumption of a linear evolution of the surface switching during switching and through geometric considerations suggest that the cycle of work due to the behavior of the surface of switching. Taking advantage of the switching frequency sets obtained with the algorithm ZAD has proposed a new realization of the interleaving technique applied to modular power systems which takes into account a variable number of converters assets. It has also devoted special attention to the design and implementation of management strategies modular operating system to achieve the benefits associated with such systems in terms of efficiency, flexibility, reliability and ease of maintenance. Another important part of this work is the realization of control strategies proposed, as well as the management system operating through programmable logic FPGA devices, as this type of components which provides better benefits in terms of cost, design flexibility, possibility of concurrent execution of algorithms and high processing speed. We present experimental and simulation results obtained in different tests with a modular system consisting of three investors deliberately desapareados. The results have revealed the stability and robustness of the whole front of load variations, conexión-desconexión modules, as well as the effectiveness of the equalization flows through modules investors achieved with controls designed, thereby validating designs proposed control mode sliding, both free frequency as fixed frequency, as well as designs proposed management of f 8 uncionam 2a2 iento the modular system.

Author: MORATAL PÉREZ DAVID.
Year: 2005.
University: POLITÉCNICA DE VALENCIA [www.upv.es].
Place of defense: Dep. Ingenieria Electronica.
Place of preparation: Universidad Politécnica de Valencia.

Summary: Speed is an important factor in many dynamic MRI applications including cardiac imaging. Respiratory artifacts are commonly avoided by acquiring all data during breath-holding. An important limiting factor in cardiac imaging is thus the breath-hold duration that the subject can sustain. For patients with severe cardiovascular disease, and even for healthy subjects, there are limits to both the breath-hold duration and the number of consecutive breath-hold scans that can be acquired. Faster acquisitions allow the same image to be acquired in a shorter breath-hold, higher-resolution images to be acquired in the same breath-hold time, and complete multislice coverage of anatomy to be completed in fewer breath-holds by scanning multiple slices in one breath-hold. When imaging time pushes the limits of the patient's breath-hold capacity, shorter scan times may actually improve the quality of the images. Reduced breath-hold times may improve patient throughput and result in more successfully completed studies. Dynamic imaging often involves repeated acquisitions of an image plane in which only part of the FOV changes over time. For example, in cardiac MRI the heart moves during the cardiac cycle, but the lungs and shoulders remain in the same location. As a rule, the dynamic parts of the image contain the interesting structures in the image. Sometimes static regions provide an anatomical reference, but often the only reason for including static regions in the FOV is to avoid contamination of relevant dynamic regions by aliasing or wraparound artifacts. Conventional dynamic MRI acquires sufficient data to reconstruct these static regions independently for each image of a time sequence. Such data are intrinsically redundant, since ideally the static region is identical in all images. In this thesis, a novel acceleration strategy called "Noquist" is presented. Using Noquist, a more sparsely sampled dynamic image sequence is reconstructed correctly without Nyquist foldover artifact by reductions in the size of the discrete Fourier model of the k-space data for the dynamic image. The image is reconstructed by direct inversion of this reduced model. Our approach does not require synthesis of omitted conventional-grid data or after-the-fact removal of foldover artifacts, and accommodates variations in the dynamic FOV fraction easily and naturally. The theory behind this method is introduced along with sample reconstructions from real and simulated data. Noquist is also compared with other existing rFOV techniques, studying their SNR behavior as well as their conduct with moving objects and structures. In principle, Noquist does not require the dynamic FOV be contiguous. This observation may be of particular interest for flow imaging, where there may be no motion of large contiguous organs, but only motion by intensity changes in several blood vessels that appear in disjoint locations in the FOV, with static regions in between. The use of Noquist reconstruction in such a situation may require a different reconstruction matrix than the contiguous case, but the data reduction is the same as for a contiguous dynamic FOV of the same accumulated size. A multiregion version of Noquist method is developed and its first results are exposed.

Author: RON ANGEVIN RICARDO.
Year: 2005.
University: MÁLAGA [www.uma.es].
Place of defense: E.T.S.I. TELECOMUNICACIÓN.
Place of preparation: E.T.S.I. TELECOMUNICACIÓN.

Summary: A front cerebro-computadora or BCI (Brain-Computer Interface), is based primarily on analysis of the signals electroencephalographic (EEG signals) captured during some kind of mental activity with the purpose of controlling an external component. Its benefits will largely depend on its ability to control a subject for their EEG patterns, adequate training is necessary in some cases can reach extended in time, and becoming imperative to provide some kind of feedback (feedback) that allows its evolution. The objective of this thesis is focused on a study on the possible consequences of using a feedback based on virtual reality techniques in training systems BCI. To make the study and validate power, it has become necessary to compare the results obtained when using these techniques with those obtained through a system BCI based reference feedback as the conventional length of a horizontal bar. This comparative study has therefore compelled to develop two different types of interfaces cerebro-computadora and ultimately perform the work in two different stages. In a first phase, has been developed and evaluated the system BCI reference allowing results that could be compared with the interface object of this work. In a second phase, has been developed and evaluated based on the system BCI virtual reality techniques. The results show a different behavior in the response of subjects achieving better control of the EEG signals, particularly during the period of feedback. It seems that the use of an interface that is more familiar and more attractive to the subject allows greater motivation and integrating them and may improve outcomes for classification, while allowing greater and more rapid adaptation to the subject of paradigm training.

Author: PALACIÓN ROCA JORGE.
Year: 2005.
University: BARCELONA [www.ub.es].
Place of defense: FACULTAT DE FÍSICA.
Place of preparation: FACULTAT DE FÍSICA.

Summary: This thesis has been addressed compact modeling of dynamic thermal behavior of components and systems. Taking the cases to analyze a microbomba, a stack of chips, a microactuador fireworks and a termopila have proposed various methods of analysis to get models with a single port heat, using non - parametric methods, and several thermal ports using parametric methods. In all cases there has been modeling the thermal performance with a high degree of agreement between the experimental temperature and simulated. The main feature of the proposal is that all the models proposed are achievable heat in the form of networks of resistors and capacitors because in this way are easily incorporables in electrical conventional simulators.

Author: BEDOYA JIMENEZ GUILLERMO.
Year: 2005.
University: POLITÉCNICA DE CATALUÑA [www.upc.edu].
Place of defense: CAMPUS NORD, EDIFICI C4, AULA 002.
Place of preparation: C4 Jordi Girona, 1-3, campus nord 08034.

Summary: This work is concerned with the problem of Nonlinear Blind Source Separation (BSS). In a more specific context, we deal with the implementation of a new class of smart sensor systems founded on the use of semiconductor-based chemical sensor arrays, interface electronics and neural-based source separation algorithms. The main goal is to obtain low cost/high performance integrated sensor system architectures. In the particular case of chemical sensors with poor performance, we propose the use of novel extended Nonlinear BSS techniques to improve their performance, give spatial selectivity and cancel cross-talks between several sources in the array. The thesis is focused on the search of the increase in efficiency and performance of smart chemical sensor systems that includes noise-tolerant and non-linear signal processing tools, with the role of allows a signal enhancement for the extraction of useful information from the sensors output. This implies researching basic questions such as what are strategies that can optimally use the information contained in the data from the statistics perspective, and the study, extension and application of particular Nonlinear BSS mixtures models. The main work is focused on the next topics: 1. Study and development of source separation algorithms based on the post nonlinear mixture model, and its application to semiconductor-based bio-chemical sensor arrays. 2. Study and development of source separation algorithms based on the bilinear mixture model, and its application to TGS gas sensor arrays systems. 3. Development of signal processing strategies to improve the Bio-chemical and Gas sensors performance, given spatial selectivity and cancelling non linearities and cross-talksbetween several species in an array. Development of advanced learning strategies oriented to separate single chemical species from multi-component mixtures in high-noise environments. 4. A global description of the algorithmic constraints and future trends. This proposal is understood as a novel alternative to the classic approaches used in chemical sensing, which have been oriented to the improvement of the physical proprieties of sensors, the classification and pattern recognition schemes, and the multivariate data analysis. This is expected to have a strong influence on real-world applications to non-linear and noise-tolerant chemical identification systems.

Author: SALAET PEREIRA JOAN FERRAN.
Year: 2005.
University: POLITÉCNICA DE CATALUÑA [www.upc.edu].
Place of defense: AULA 28.8 PLANTA 1, PABELLÓ 1- ETSEIB, CAMPUS SUD.
Place of preparation: C4 Jordi Girona, 1-3, campus nord 08034.

Author: MARFIL ROBLES REBECA.
Year: 2005.
University: MÁLAGA [www.uma.es].
Place of defense: E.T.S.I. TELECOMUNICACIÓN UNIVERSIDAD DE MÁLAGA.
Place of preparation: E.T.S.I. TELECOMUNICACIÓN. UNIVERSIDAD DE MÁLAGA.

Summary: This thesis proposes a system for tracking of objects based on a new method of representation and location of the target. This track non-rigid objects in real time, without using any previous form of objects forward. To achieve this, we propose a new model for characterizing the appearance of the object based on a mask or template. This model uses a new pyramid structure called Bounded Irregular Pyramid (BIP) to represent the target and template, as well as to make the process of locating the object or template matching hierarchical manner, reducing its computational cost. The proposed tracking system allows tracking of rigid and non-rigid objects using a mask that is updated dynamically. Esta máscara incluye información de las máscaras previas, solucionando dos de las causas de fallo más importante se los sistemas de seguimiento: cambios en la apariencia del objetivo y oclusiones del mismo. In addition, the system allows tracking multiple objects simultaneously without an excessive increase in the computational cost. The features previously commented on the proposed system makes it very suitable for use in applications visually more complex, requiring a response in real time.

Author: SOTO VALLES FULGENCIO.
Year: 2005.
University: POLITÉCNICA DE CARTAGENA [www.upct.es].
Place of defense: UNIVERSIDAD POLITÉCNICA DE CARTAGENA.
Place of preparation: UNIVERSIDAD POLITÉCNICA DE CARTAGENA.

Summary: With the increase in economic importance for bluefin tuna (Thunnus Thynnus) and the need to supply the growing demand for the consumer market, the number of companies engaged in the industrial exploitation of the same through cage culture, is growing. This thesis proposes new slaughter techniques that solve problems arising from the use of existing in this sector. The main problem with the current slaughter techniques, is the high level of stress they are subjected to tuna, which causes a decline in the quality of meat. The proposed technique is the use of electricity to get the electronarcosis and ultimately the electrosacrificio of tuna, as it is being used with great success with small freshwater fish for nearly a century. This method is selected because they sacrifice the advantages of being selective, minimally stressful for the fish, and in accordance with the existing rules and recommendations on animal welfare in the slaughter. To apply this method is carried out an exhaustive study of the technique electropesca freshwater, to evaluate its adaptation to the case study. Given the high conductivity difference between freshwater and the water settles, it is preferable to implement downloading contact (using a harpoon), and not through dela creating an electric field as is done in freshwater. With this technique is conducting a study comparing the incidence of use of different waveforms on the quality of the meat obtained quality is judged by two parameters, Core or color of the meat, and RC or spinal damage. The waveforms are tested; continuous flow, flow pulsatoria high and low frequency, pulse complex system (CPS), AC regulated and a mixed signal. Some of them had already been used in freshwater, although not for others. The results of the tests performed, it appears that the waveforms widely used in the electropesca in freshwater, only the DC pulsatoria, has also been effective pair wing electropesca water settles. However, other waveforms not tested in fresh water, have been more effective. Specifically mixed signal, mixing DC pulsatoria and alternating current, has been tested all considered the most appropriate, since in addition to maintaining an acceptable quality level requires some moderate values of tension. In view of these findings, there is a need to do more tests and other waveforms totally different. It appears above the need for a flexible team, which can be programmed waveforms and record data associated with each signal. It designs and implements therefore a multifunctional and programmable electronic equipment for use in aquaculture. It shows each of the parties that make up the equipment, hardware, software and the necessary interface between them. It details the steps for the design and implementation of hardware, as well as the peculiarities which must be met for the proposed implementation, where the power converter is subjected to constant shorts and where it is intended that the applied voltage at discharge is as similar possible to the desired. Similarly reflects the experimental results obtained with the team to complete the tests on land.

Author: COSTAS PÉREZ LUCÍA.
Year: 2005.
University: VIGO [www.uvigo.es].
Place of defense: E.T.S. INGENIEROS INDUSTRIALES.
Place of preparation: E.T.S. INGENIEROS INDUSTRIALES.

Summary: The purpose of this dissertation is to develop a methodology to provide efficient mechanisms denouncement concurrent error allowing raise the reliability of complex digital systems. The new methodology has been called detection algoritmica and has been successfully applied to a system of fixed image compression based on wavelet transform two-dimensional. Because the validation tools available are not the most appropriate in this particular case, have developed two new environments for this purpose. The first one is based on logic simulation and emulation in the second through devices configurable (FPGAS). Both have been used to validate the proposed criteria for detection.

Author: NESCOLARDE SELVA LEXA DIGNA.
Year: 2005.
University: POLITÉCNICA DE CATALUÑA [www.upc.edu].
Place of defense: AULA DE TELEENSENYAMENT - ETSETB - CAMPUS NORD.
Place of preparation: EDIFICI U DESPATX 518 CAMPUS SUD.

Author: Maksymov Ivan.
Year: 2006.
University: ROVIRA I VIRGILI [www.urv.cat].
Place of defense: Escola Tècnica Superior d'Enginyeria.
Place of preparation: ETSE.

Summary: In this dissertation was a study of various properties of the crystals fotónicos 1D and 2D nonlinear third order and how these properties can be applied to the development of fully optical devices (such as limiters and switches, logic gates, transistors optical, etc..). It proposed a numerical approximation to calculate the basic characteristics of crystals fotónicos nonlinear, such as the diagram of bands or transmission. The numerical approximation presented in the thesis has certain advantages useful for anyone who designs optical devices based on crystals fotónicos nonlinear. The software developed based on this approach has enabled digital design and numerically simulate an optical switch entirely whose benefits are superior to those of conventional optoelectronic devices.

Author: SOSA AVENDAÑO JORGE LUIS.
Year: 2006.
University: POLITÉCNICA DE CATALUÑA [www.upc.edu].
Place of defense: Sala de actos de EPSEVG.
Place of preparation: C4 Jordi Girona, 1-3, campus nord 08034.

Summary: This work proposes a dynamic study for DC-DC resonant converters operating with the most novel modulation techniques that allow to extend the zero voltage switch (ZVS) range in this type of converters. The study includes the synthesis of a suitable averaged large-signal dynamic model for each topology and the design of nonlinear feedback controllers based in the input-output linearization approach. The modeling method is based on the extended describing functions, which approximate the resonant state-variables to sinusoidal functions with slow time-varying amplitude and phases. The proposed controllers provide some outstanding features as: ZVS condition for a wide load range, fast transient response and robustness in relation to external parameter variations. The work is organized as follows: Chapter 2 presents the different modulation method that provides zero voltage switch in a wider dynamic range. Also the steady-state characteristics based on the first harmonic approximation of each resonant circuit are presented. In the chapter 3 the modeling method is applied to three resonant topologies (series, CLL-T and series-parallel LCC resonant converter) to obtain its averaged large-signal models. In order to verify the model prediction capabilities, a comparison of the resonant variables waveforms generated by the averaged models and the state space models has been carried out. Chapter 4 describes the design of nonlinear controllers for each resonant circuit and modulation technique based on the large-signal models obtained in chapter 3. The proposed controllers use linear sliding surfaces obtained by dynamic imposition, which allow the converters to follow a linear dynamic response when they are operating in closed-loop. Chapter 5 presents the results obtained with experimental prototypes which confirm the validity of the models and the proposed controllers. Finally, the conclusions of the work and the future investigation lines are presented in the chapter 6.

Author: BARRAGAN MENDOZA NELSON ALFREDO.
Year: 2006.
University: OVIEDO [www.uniovi.es].
Place of defense: E.P.S. DE INGENIEROS INDUSTRIALES.
Place of preparation: E.P.SUPERIOR DE INGENIEROS INDUSTRIALES, INFORMATICOS Y TELECOMUNICACIONES GIJON (SALA DE JUNTAS).

Summary: The first part of this Doctoral Thesis is art that begins with a brief tour of the phenomena directly involved in the development of fiber optic chemical sensors, then describes the current status of the fiber-optic sensors. The core of the thesis is the following chapters in the developing work on the structure of generation of knowledge-based process Analysis, Synthesis and Comparison of solutions. In this line, first looks at the various options available for solving the specific problems of the excitement of phosphors as sensors to gather the low levels of light that characterize these sensors. For each of the various possibilities have been synthesized solutions optimized both solid state and capable of miniaturization, as more classical solutions based on the use of arc lamps or PMTs high sensitivity. Each of the solutions has been tested experimentally so extensive, both seeking to determine their characteristics instrumental as his margin of use and implementation in the case of chemical sensors phosphors. Finally, as a result of the experimental results have been able to establish criteria for determining the scope of application of each and the advantages and disadvantages they present, highlighting their parameterization optimized. The work concludes with a summary of the input and avenues of study that arise from developed so far.