The low complexity of IIR adaptive filters (AFs) is specially appealing to realtime applications but some drawbacks have been preventing their widespread use so far. For gradient based IIR AFs, adverse operational conditions cause convergence problems in system identification scenarios: underdamped and clustered poles, undermodelling or non-white input signals lead to error surfaces where the adaptation nearly stops on large plateaus or get stuck at sub-optimal local minima that can not be identified as such a priori. Furthermore, the non-stationarity in the input regressor brought by the filter recursivity and the approximations made by the update rules of the stochastic gradient algorithms constrain the learning step size to small values, causing slow convergence. In this work, we propose IIR performance enhancement strategies based on hybrid combinations of AFs that achieve higher convergence rates than ordinary IIR AFs while keeping the stability. / A baixa complexidade dos filtros adaptativos (FAs) IIR é atrativa para aplicações em tempo real, mas certos inconvenientes têm impedido sua ampla utilização até agora. Para os FAs baseados no gradiente descendente, condições operacionais adversas suscitam problemas de convergência em cenários de identificação de sistemas: pólos subamortecidos ou agrupados, submodelagem ou sinais correlacionados originam superfícies de erro onde a adaptação desacelera em grandes planícies ou para em mínimos locais sub-ótimos que não podem ser identificados como tais a priori. Além disso, a não-estacionaridade do regressor de entrada causada pela recursividade do filtro e as aproximações feitas pelas regras de atualização dos algoritmos de gradiente estocástico restringem o passo de aprendizado a valores pequenos, retardando a convergência. Neste trabalho, propomos estratégias de aprimoramento de desempenho baseadas em combinações híbridas e estáveis de FAs que alcançam taxas de convergências mais altas do que FAs IIR comuns.

The trend seasonal demand pattern is encountered when both trend and seasonal influences are interactive. The problem of this research is to project the seasonal market sales using ice cream and fresh milk in Taiwan as examples. In order to improve the accuracy of forecast, two different methods are validated and the best forecasting method is selected based on the minimum Mean Square Error. In this study, we present two forecasting models used for evaluation to predict seasonal market sales of ice cream, fresh milk, and air conditioner in Taiwan. It includes Winters multiplicative seasonal trend model and the Decomposition method. Two different methods are validated and the best forecasting method is selected based on the minimum Mean Square Error. After the validation process, Winters multiplicative seasonal trend model is selected based on the minimum MSE, and the monthly sales forecast for the year of 2011 is conducted using the data(60 months). Number Cruncher Statistical System (NCSS) is used for analyzing the data which proves useful and powerful. In summary, the results demonstrate that Winters multiplicative seasonal trend model has the smallest mean square error in this case. Therefore, we conclude that both Winters multiplicative seasonal trend model and the Decomposition model are well fitted for forecasting the seasonal market sales. Yet, Winters multiplicative seasonal trend model is the better method to be used in this study since it generates the smallest mean square error (MSE) during the period of validation.

The efficiency of direct sequence spread spectrum (DS/SS) receiver is highly dependent on the accurate and fast synchronization between the incoming and locally generated PN (pseudo-noise) codes. The code synchronization is processed in two steps, acquisition (coarse alignment) and tracking (fine alignment), to estimate the delay offset between the two codes. In general, the schemes for code acquisition and tracking processes are performed, separately, and implemented with different structure. Recently, an alternative approach, with the adaptive LMS filtering scheme, has been proposed for performing both code acquisition and tracking with the identical structure, where the coherent receiver was employed. With this approach, dramatically, hardware complexity reduction could be achieved, especially, when long PN code is considered. In this thesis, a new differentially coherent code synchronization scheme, based on a differential detector followed by an adaptive constrained LMS (CLMS) filtering algorithm with maximum tap weight (MTW) test scheme, is devised for performing both code acquisition and tracking with the identical structure. With a differential detector for code synchronization, the prior knowledge of the carrier phase is not required as the non-coherent techniques. Numerical analyses and simulation results verify that the proposed scheme has better acquisition performance, in terms of mean acquisition time, than the conventional LMS filtering algorithm with MTW test and mean square error (MSE) test schemes for the integer and non-integer time delay environments. At the same time, the proposed scheme has better tracking capability, in terms of mean hold-in time and mean penalty time, over the conventional LMS filtering schemes, for the variation of signal-to-noise ratio (SNR) and delay offset (delay difference).

Effort to estimate unknown time-varying channels as a part of high-speed mobile communication systems is of interest especially for next-generation wireless systems. The high computational complexity of the optimal Wiener estimator usually makes its use impractical in fast time-varying channels. As a powerful candidate, the adaptive least mean squares (LMS) algorithm offers a computationally efficient solution with its simple first-order weight-vector update equation. However, the performance of the LMS algorithm deteriorates in time-varying channels as a result of the eigenvalue disparity, i.e., spread, of the input correlation matrix in such chan nels. In this work, we incorporate the L MS algorithm into the well-known bidirectional processing idea to produce an extension called the bidirectional LMS. This algorithm is shown to be robust to the adverse effects of time-varying channels such as large eigenvalue spread. The associated tracking performance is observed to be very close to that of the optimal Wiener filter in many cases and the bidirectional LMS algorithm is therefore referred to as near-optimal. The computational complexity is observed to increase by the bidirectional employment of the LMS algorithm, but nevertheless is significantly lower than that of the optimal Wiener filter. The tracking behavior of the bidirectional LMS algorithm is also analyzed and eventually a steady-state step-size dependent mean square error (MSE) expression is derived for single antenna flat-fading channels with various correlation properties. The aforementioned analysis is then generalized to include single-antenna frequency-selective channels where the so-called ind ependence assumption is no more applicable due to the channel memory at hand, and then to multi-antenna flat-fading channels. The optimal selection of the step-size values is also presented using the results of the MSE analysis. The numerical evaluations show a very good match between the theoretical and the experimental results under various scenarios. The tracking analysis of the bidirectional LMS algorithm is believed to be novel in the sense that although there are several works in the literature on the bidirectional estimation, none of them provides a theoretical analysis on the underlying estimators. An iterative channel estimation scheme is also presented as a more realistic application for each of the estimation algorithms and the channel models under consideration. As a result, the bidirectional LMS algorithm is observed to be very successful for this real-life application with its increased but still practical level of complexity, the near-optimal tracking performa nce and robustness to the imperfect initialization.

We study anomalous diffusion on fractals with a static external field applied. We utilise the master equation to calculate particle distributions and from that important quantities as for example the mean square displacement <r^2(t)>. Applying different bias amplitudes on several regular Sierpinski carpets we obtain maximal drift velocities for weak field strengths. According to <r^2(t)>~t^(2/d_w), we determine random walk dimensions of d_w<2 for applied external fields. These d_w corresponds to superdiffusion, although diffusion is hindered by the structure of the carpet, containing dangling ends. This seems to result from two competing effects arising within an external field. Though the particles prefer to move along the biased direction, some particles get trapped by dangling ends. To escape from there they have to move against the field direction. Due to the by the bias accelerated particles and the trapped ones the probability distribution gets wider and thus d_w<2. / In dieser Arbeit untersuchen wir anomale Diffusion auf Fraktalen unter Einwirkung eines statisches äußeres Feldes. Wir benutzen die Mastergleichung, um die Wahrscheinlichkeitsverteilung der Teilchen zu berechnen, um daraus wichtige Größen wie das mittlere Abstandsquadrat <r^2(t)> zu bestimmen. Wir wenden unterschiedliche Feldstärken bei verschiedenen regelmäßigen Sierpinski-Teppichen an und erhalten maximale Driftgeschwindigkeiten für schwache Feldstärken. Über <r^2(t)>~t^{2/d_w} bestimmen wir die Random-Walk-Dimension d_w als d_w<2. Dieser Wert für d_w entspricht der Superdiffusion, obwohl der Diffusionsprozess durch Strukturen des Teppichs, wie Sackgassen, behindert wird. Es schient, dass dies das Ergebnis zweier konkurrierender Effekte ist, die durch das Anlegen eines äußeren Feldes entstehen. Einerseits bewegen sich die Teilchen bevorzugt entlang der Feldrichtung. Andererseits gelangen einige Teilchen in Sackgassen. Um die Sackgassen, die in Feldrichtung liegen, zu verlassen, müssen sich die Teilchen entgegen der Feldrichtung bewegen. Somit sind die Teilchen eine gewisse Zeit in der Sackgasse gefangen. Infolge der durch das äußere Feld beschleunigten und der gefangenen Teilchen, verbreitert sich die Wahrscheinlichkeitsverteilung der Teilchen und somit ist d_w<2.

This manuscript concerns the performance analysis in signal processing and consists into two parts : First, we study the lower bounds in characterizing and predicting the estimation performance in terms of mean square error (MSE). The lower bounds on the MSE give the minimum variance that an estimator can expect to achieve and it can be divided into two categories depending on the parameter assumption: the so-called deterministic bounds dealing with the deterministic unknown parameters, and the so-called Bayesian bounds dealing with the random unknown parameter. Particularly, we derive the closed-form expressions of the lower bounds for two applications in two different fields: (i) The first one is the target localization using the multiple-input multiple-output (MIMO) radar in which we derive the lower bounds in the contexts with and without modeling errors, respectively. (ii) The other one is the pulse phase estimation of X-ray pulsars which is a potential solution for autonomous deep space navigation. In this application, we show the potential universality of lower bounds to tackle problems with parameterized probability density function (pdf) different from classical Gaussian pdf since in X-ray pulse phase estimation, observations are modeled with a Poisson distribution. Second, we study the statistical resolution limit (SRL) which is the minimal distance in terms of the parameter of interest between two signals allowing to correctly separate/estimate the parameters of interest. More precisely, we derive the SRL in two contexts: array processing and MIMO radar by using two approaches based on the estimation theory and information theory. We also present in this thesis the usefulness of SRL in optimizing the array system.

No description available.

Single-particle tracking (SPT) is a method used to study the diffusion of various molecules within the cell. SPT involves tagging proteins with optical labels and observing their individual two-dimensional trajectories with a microscope. The analysis of this data provides important information about protein movement and mechanism, and is used to create multistate biological models. One of the challenges in SPT analysis is the variety of complex environments that contribute to heterogeneity within movement paths. In this thesis, we explore the limitations of current methods used to analyze molecular movement, and adapt analytical methods used in animal movement analysis, such as correlated random walks and first-passage time variance, to SPT data of leukocyte function-associated antigen-1 (LFA-1) integral membrane proteins. We discuss the consequences of these methods in understanding different types of heterogeneity in protein movement behaviour, and provide support to results from current experimental work. / Applied Mathematics

Made available in DSpace on 2015-03-26T13:32:05Z (GMT). No. of bitstreams: 1 texto completo.pdf: 688077 bytes, checksum: 369ec0145d58b4c3f2d93ab69403df95 (MD5) Previous issue date: 2009-01-19 / Coordenação de Aperfeiçoamento de Pessoal de Nível Superior / Many times, the interpretation of the results in cluster analysis is done subjectively, that is, through inspection on dendograms, since there are no objective criteria to identify the formed clusters. In face of such a problem, the present study aimed to: (1) find out an objective way to achieve the cut-point (optimal number of clusters) in a dendogram in order to help on taking the right decision; (2) work out index concepts such as Root Mean Square Standard Deviation (RMSSTD) and R Squared (RS), explaining the contribution of each one of them in determining the optimal number of cluster; (3) method application, aiming to identify divergent accessions that will be used on improvement programs. An alternative solution for this problem is to use the RMSSTD and RS which are calculated according to the number of variables among and within the clusters formed, characterizing an objective way to determine the optimal number. Another solution is achieved by using the RS. Some morphological characteristics of the forty nine accessions of the species Capsicum chinense Jacq. from the Germplasm Bank of Vegetables of the Federal University of Viçosa (Banco de Germoplasma de Hortaliças da Universidade Federal de Viçosa, Minas Gerais Brazil) were analyzed by means of cluster analysis. The accessions were clustered based on the proposed techniques and an optimal number of clusters was achieved. The 49 accessions analyzed were classified into only seven clusters according to the graph of the RMSSTD versus the number of clusters and the graph of the RS versus the number of clusters. / Muitas vezes, a interpretação dos resultados em análise de agrupamentos é feita de forma subjetiva, isto é, através da inspeção de dendrogramas. Isto se deve ao fato de haver dificuldade em se encontrar na literatura um critério objetivo de fácil aplicação para identificar o número ideal de grupos formados. Diante deste problema, o presente trabalho teve por objetivos: 1) Avaliar a aplicabilidade de critério objetivo de se obter o ponto de corte (número ótimo de clusters) num dendrograma para a tomada de decisão; 2) trabalhar os conceitos de índices como RMSSTD (root mean square standard deviation) e RS (R-Squared), discutindo a contribuição de cada um destes na obtenção do número ótimo de clusters em acessos de Capsicum chinense; 3) aplicação do método, visando a identificar acessos divergentes de Capsicum chinense para serem utilizados em programas de melhoramento. Os índices RMSSTD e RS são calculados de acordo com as variáveis entre e dentro dos grupos formados, caracterizando uma forma objetiva para determinar o número ótimo. Para se obter o ponto de máxima curvatura da trajetória dos índices RMSSTD e RS em função do aumento do número de grupos (X), utilizou-se o Método da Máxima Curvatura Modificado. Foram analisadas, por meio da análise de agrupamentos, algumas características morfológicas de quarenta e nove acessos da espécie Capsicum chinense Jacq. do Banco de Germoplasma de Hortaliças da Universidade Federal de Viçosa. A partir das técnicas propostas agrupou-se os acessos, obtendo um número ótimo de grupos. Os resultados classificam os 49 acessos avaliados em apenas sete grupos de acordo com o gráfico do RMSSTD versus o número de grupos e o gráfico do RS versus o número de grupos.

Submitted by Automação e Estatística (sst@bczm.ufrn.br) on 2018-04-11T20:06:38Z No. of bitstreams: 1 BrunoJacomeCavalcanti_TESE.pdf: 5397909 bytes, checksum: 5a245eec570a69adf8ca5d791aaddf70 (MD5) / Approved for entry into archive by Arlan Eloi Leite Silva (eloihistoriador@yahoo.com.br) on 2018-04-16T20:46:19Z (GMT) No. of bitstreams: 1 BrunoJacomeCavalcanti_TESE.pdf: 5397909 bytes, checksum: 5a245eec570a69adf8ca5d791aaddf70 (MD5) / Made available in DSpace on 2018-04-16T20:46:19Z (GMT). No. of bitstreams: 1 BrunoJacomeCavalcanti_TESE.pdf: 5397909 bytes, checksum: 5a245eec570a69adf8ca5d791aaddf70 (MD5) Previous issue date: 2017-10-20 / A perfeita funcionalidade dos sistemas de comunicações de 3ª. e 4ª. gerações requerem, entre outras coisas, do conhecimento dos valores numéricos da predição das perdas de propagação dos sinais propagantes nos ambientes urbano, suburbano e rural. Portanto, o estudo das condições de propagação em um ambiente qualquer sempre será uma preocupação dos engenheiros projetistas. A análise e desenvolvimento de modelos robustos de predição de perdas de propagação em redes de comunicações Long Term Evolution (LTE) e Long Term Evolution Advanced (LTE-A) usando técnicas de Inteligência Artificial são realizadas neste trabalho. Os procedimentos metodológicos empregados foram aplicados no melhoramento da predição dos modelos de perda de propagação empíricos SUI, ECC-33, Ericsson 9999, TR 36.942 e o modelo do Espaço Livre, aplicados em redes LTE e LTE-A nas frequências de 800 MHz, 1800 MHz e 2600 MHz, para ambientes suburbanos em cidades de porte médio do nordeste do Brasil. Assim, nesta tese propõem-se dois modelos de Redes Neurais Artificiais (RNA): (i) o modelo de rede neural com entradas baseadas em erro (RNBE), utilizando como principal alimentador da rede o erro entre dados medidos e simulados, e, (ii) o modelo de rede neural com entradas baseadas no terreno (RNBT). O desempenho desses modelos foram comparados com os modelos de propagação considerados no trabalho e também as versões otimizadas utilizando Algoritmos Genéticos (AG) e o Método dos Mínimos Quadrados (LMS). Também foram realizadas comparações com valores medidos, obtidos a partir de uma campanha de medição realizada na cidade de Natal, Estado do Rio Grande do Norte. Os resultados finais obtidos através de simulações e medições apresentaram boas concordâncias métricas, com destaque para a performance do modelo RNBE. A principal contribuição dessa tese é que, ao utilizar essas técnicas que fazem uso de maneira mais eficiente dos modelos de propagação empíricos, pode-se estimar sinais propagantes realistas, evitando erros no planejamento e implementações de redes sem fio LTE e LTE-A em áreas suburbanas. / The perfect functionality of the 3rd and 4th generation of wireless systems requires, among other parameters, knowledge of the numerical values of the prediction of loss of propagation of propagation signals in urban, suburban and rural environments. Therefore, the study of propagation conditions in any environment will always be a concern of design engineers. The analysis and development of robust propagation loss prediction models in Long Term Evolution (LTE) and Long Term Evolution Advanced (LTE-A) communications networks using Artificial Intelligence techniques is performed in this work. The methodologies used were applied to improve the prediction of loss of empirical propagation SUI, ECC-33, Ericsson 9999, TR 36.942 models and the Free Space model applied in LTE and LTE-A networks in the frequencies of 800 MHz, 1800 MHz and 2600 MHz, for suburban environments in mid-sized cities in northeastern Brazil. Thus, in these thesis two models of Artificial Neural Networks (RNA) are proposed: (i) the neural network model with inputs based on error (RNBE) using as main feeder of the network the error between measured and simulated data, and (ii) the neural network model with land-based inputs (RNBT). The performance of these models was compared with the models of propagation considered in the work and also the versions optimized using Genetic Algorithms (AG) and the Least Square Method (LMS). Comparisons were also made with measured values, obtained from a measurement campaign carried out in the city of Natal, state of Rio Grande do Norte. The final results obtained through simulations and measurements presented good metric concordances, with emphasis on the performance of the RNBE model. Thus, the main contribution of this thesis is that, by using these techniques that make more efficient use of empirical propagation models, we can estimate realistic propagation signals, avoiding errors in the planning and implementations of LTE and LTE- A wireless networks in suburban areas.