A novel parallel approach to the likelihood-based estimation of admixture in population genetics

Inferring population admixture from genetic data and quantifying it is a difficult but crucial task in evolutionary and conservation biology. Unfortunately state-of-the-art probabilistic approaches are computationally demanding. Effectively exploiting the computational power of modern multiprocessor systems can thus have a positive impact to Monte Carlo-based simulation of admixture modeling. A novel parallel approach is briefly described and promising results on its message passing interface (MPI)-based C implementation are reported.

Spectroscopy of highly excited vibrational states of HCN in its ground electronic state

An experimental technique based on a scheme of vibrationally mediated photodissociation has been developed and applied to the spectroscopic study of highly excited vibrational states in HCN, with energies between 29 000 and 30 000 cm(-1). The technique consists of four sequential steps: in the first one, a high power laser is used to vibrationally excite the sample to an intermediate state, typically (0,0,4), the nu(3) mode being approximately equivalent to the C-H stretching vibration. Then a second laser is used to search for transitions between this intermediate state and highly vibrationally excited states. When one of these transitions is found, HCN molecules are transferred to a highly excited vibrational state. Third, a ultraviolet laser photodissociates the highly excited molecules to produce H and CN radicals in its A (2)Pi electronic state. Finally, a fourth laser (probe) detects the presence of the CN(A) photofragments by means of an A-->B-->X laser induced fluorescence scheme. The spectra obtained with this technique, consisting of several rotationally resolved vibrational bands, have been analyzed. The positions and rotational parameters of the states observed are presented and compared with the results of a state-of-the-art variational calculation. (C) 2004 American Institute of Physics.

Experimental model of the interfacial instability in aluminium reduction cells

A solution has been found to the long-standing problem of experimental modelling of the interfacial instability in aluminium reduction cells. The idea is to replace the electrolyte overlaying molten aluminium with a mesh of thin rods supplying current down directly into the liquid metal layer. This eliminates electrolysis altogether and all the problems associated with it, such as high temperature, chemical aggressiveness of media, products of electrolysis, the necessity for electrolyte renewal, high power demands, etc. The result is a room temperature, versatile laboratory model which simulates Sele-type, rolling pad interfacial instability. Our new, safe laboratory model enables detailed experimental investigations to test the existing theoretical models for the first time.

Interaction of nitrogen dioxide (NO2) with a monolayer of oleic acid at the air–water interface – A simple proxy for atmospheric aerosol

The reactions between atmospheric oxidants and organic amphiphiles at the air water interface of an aerosol droplet may affect the size and critical supersaturation required for cloud droplet formation. We demonstrate that no reaction occurs between gaseous nitrogen dioxide (1000 ppm in air) and a monolayer of an insoluble amphiphile, oleic acid (cis-9-octadecenoic acid), at the air water interface which removes material from the air water interface. We present evidence that the NO2 isomerises the cis-9-octadecenoic (oleic) acid to trans-9-octadecenoic (elaidic) acid. The study presented here is important for future and previous studies of (1) the reaction between the nitrate radical, NO3, and thin organic films as NO2 is usually present in high concentrations in these experimental systems and (2) the effect of NO2 air pollution on the unsaturated fatty acids and lipids found at the air liquid surface of human lung lining fluid.

B-spline neural network based digital baseband predistorter solution using the inverse of De Boor algorithm

In this paper a new nonlinear digital baseband predistorter design is introduced based on direct learning, together with a new Wiener system modeling approach for the high power amplifiers (HPA) based on the B-spline neural network. The contribution is twofold. Firstly, by assuming that the nonlinearity in the HPA is mainly dependent on the input signal amplitude the complex valued nonlinear static function is represented by two real valued B-spline neural networks, one for the amplitude distortion and another for the phase shift. The Gauss-Newton algorithm is applied for the parameter estimation, in which the De Boor recursion is employed to calculate both the B-spline curve and the first order derivatives. Secondly, we derive the predistorter algorithm calculating the inverse of the complex valued nonlinear static function according to B-spline neural network based Wiener models. The inverse of the amplitude and phase shift distortion are then computed and compensated using the identified phase shift model. Numerical examples have been employed to demonstrate the efficacy of the proposed approaches.

Culture, cognition and e-commerce behavior

Existing research has given little attention to the relationship between culture characteristics and consumer’s internal beliefs particularly in the pre-purchase stage, and how this relationship affects consumer’s purchase decision. This paper considers the theory of cognitive dissonance and its extended model (the 3D-RAB), as a means to study the current distribution of consumer’s pre-purchase cognitive dissonance, which allows us to investigate the effects of culture characteristics on this distribution. Results revealed that individualism versus collectivism and high power distance dimensions, from Hofstede’s cultural model, influence consumer’s pre-purchase cognitive dissonance. These dimensions must be considered in the design of e-commerce website, by tailoring motivational/influences methods and techniques to reflect targeted consumers culture.

Non-uniform data distribution for communication-efficient parallel clustering

Global communicationrequirements andloadimbalanceof someparalleldataminingalgorithms arethe major obstacles to exploitthe computational power of large-scale systems. This work investigates how non-uniform data distributions can be exploited to remove the global communication requirement and to reduce the communication costin parallel data mining algorithms and, in particular, in the k-means algorithm for cluster analysis. In the straightforward parallel formulation of the k-means algorithm, data and computation loads are uniformly distributed over the processing nodes. This approach has excellent load balancing characteristics that may suggest it could scale up to large and extreme-scale parallel computing systems. However, at each iteration step the algorithm requires a global reduction operationwhichhinders thescalabilityoftheapproach.Thisworkstudiesadifferentparallelformulation of the algorithm where the requirement of global communication is removed, while maintaining the same deterministic nature ofthe centralised algorithm. The proposed approach exploits a non-uniform data distribution which can be either found in real-world distributed applications or can be induced by means ofmulti-dimensional binary searchtrees. The approachcanalso be extended to accommodate an approximation error which allows a further reduction ofthe communication costs. The effectiveness of the exact and approximate methods has been tested in a parallel computing system with 64 processors and in simulations with 1024 processing element

Efficient group communication for large-scale parallel clustering

Global communication requirements and load imbalance of some parallel data mining algorithms are the major obstacles to exploit the computational power of large-scale systems. This work investigates how non-uniform data distributions can be exploited to remove the global communication requirement and to reduce the communication cost in iterative parallel data mining algorithms. In particular, the analysis focuses on one of the most influential and popular data mining methods, the k-means algorithm for cluster analysis. The straightforward parallel formulation of the k-means algorithm requires a global reduction operation at each iteration step, which hinders its scalability. This work studies a different parallel formulation of the algorithm where the requirement of global communication can be relaxed while still providing the exact solution of the centralised k-means algorithm. The proposed approach exploits a non-uniform data distribution which can be either found in real world distributed applications or can be induced by means of multi-dimensional binary search trees. The approach can also be extended to accommodate an approximation error which allows a further reduction of the communication costs.

B-spline neural network based single-carrier frequency domain equalisation for Hammerstein channels

A practical single-carrier (SC) block transmission with frequency domain equalisation (FDE) system can generally be modelled by the Hammerstein system that includes the nonlinear distortion effects of the high power amplifier (HPA) at transmitter. For such Hammerstein channels, the standard SC-FDE scheme no longer works. We propose a novel Bspline neural network based nonlinear SC-FDE scheme for Hammerstein channels. In particular, we model the nonlinear HPA, which represents the complex-valued static nonlinearity of the Hammerstein channel, by two real-valued B-spline neural networks, one for modelling the nonlinear amplitude response of the HPA and the other for the nonlinear phase response of the HPA. We then develop an efficient alternating least squares algorithm for estimating the parameters of the Hammerstein channel, including the channel impulse response coefficients and the parameters of the two B-spline models. Moreover, we also use another real-valued B-spline neural network to model the inversion of the HPA’s nonlinear amplitude response, and the parameters of this inverting B-spline model can be estimated using the standard least squares algorithm based on the pseudo training data obtained as a byproduct of the Hammerstein channel identification. Equalisation of the SC Hammerstein channel can then be accomplished by the usual one-tap linear equalisation in frequency domain as well as the inverse Bspline neural network model obtained in time domain. The effectiveness of our nonlinear SC-FDE scheme for Hammerstein channels is demonstrated in a simulation study.

Two-way CSI-assisted AF relaying with HPA nonlinearity

In this paper, we investigate half-duplex two-way dual-hop channel state information (CSI)-assisted amplify-and-forward (AF) relaying in the presence of high-power amplifier (HPA) nonlinearity at relays. The expression for the end-to-end signal-to-noise ratio (SNR) is derived as per the modified system model by taking into account the interference caused by relaying scheme and HPA nonlinearity. The system performance of the considered relaying network is evaluated in terms of average symbol error probability (SEP) in Nakagami-$m$ fading channels, by making use of the moment-generating function (MGF) approach. Numerical results are provided and show the effects of several parameters, such as quadrature amplitude modulation (QAM) order, number of relays, HPA parameters, and Nakagami parameter, on performance.

On the preparation of clean tungsten single crystals

The cleaning procedure consists of two-step-flashing: (i) cycles of low power flashes T similar to 1200 K) at an oxygen partial pressure of P(o2) = 6 x 10(-8) mbar, to remove the carbon from the surface, and (ii) a single high power flash (T similar to 2200 K), to remove the oxide layer. The removal of carbon from the surface through the chemical reaction with oxygen during low power flash cycles is monitored by thermal desorption spectroscopy. The exposure to O(2) leads to the oxidation of the W surface. Using a high power flash, the volatile W-oxides and the atomic oxygen are desorbed, leaving a clean crystal surface at the end of procedure. The method may also be used for cleaning other refractory metals like Mo, Re and It. (C) 2009 Elsevier B.V. All rights reserved.

Er : YAB nanoparticles and vitreous thin films by the polymeric precursor method

The synthesis of Y(0.9)Er(0.1)Al(3)(BO(3))(4) crystalline powders and vitreous thin films were studied. Precursor solutions were obtained using a modified polymeric precursor method using D-sorbitol as complexant agent. The chemical reactions were described. Y(0.)9Er(0.1)Al(3)(BO(3))(4) composition presents good thermal stability with regard to crystallization. The Y(0.9)Er(0.1)Al(3)(BO(3))(4) crystallized phase can be obtained at 1,150 degrees C, in agreement with other authors. Crack- and porosity-free films were obtained with very small grain size and low RMS roughness. The films thickness revealed to be linearly dependent on precursor solution viscosity, being the value of 25 mPa s useful to prepare high-quality amorphous multi-layers (up to similar to 800 nm) at 740 degrees C during 2 h onto silica substrates by spin coating with a gyrset technology.

Energy transfer processes in Yb(3+)-Tm(3+) co-doped sodium alumino-phosphate glasses with improved 1.8 mu m emission

Sodium alumino-phosphate glasses co-doped with Yb(3+) and Tm(3+) ions have been prepared with notably low OH(-) content, and characterized from the viewpoint of their spectroscopic properties. In these glasses, Yb(3+) acts as an efficient sensitizer of excitation energy at 0.98 mu m - which can be provided by high power and low cost diode lasers, and subsequently undergoes non-resonant energy transfer to Tm(3+) ions ((2)F(5/2), (3)H(6) --> (2)F(7/2), (3)H(5)). Through this process, the emitting level (3)F(4) is rapidly populated, generating improved emission at 1.8 mu m ((3)F(4) --> (3)H(6)). In order to guarantee the efficiency of such favorable energy transfer, energy losses via multiphonon decay, Yb-Yb radiative trapping, and non- radiative transfer to OH(-) groups were evaluated, and minimized when possible. The dipole - dipole energy transfer microscopic parameters corresponding to Yb(3+) --> Tm(3+), Yb(3+) --> Yb(3+) and Tm(3+) --> Tm(3+) transfers, calculated by the Forster-Dexter model, are C(Yb-Tm) = 2.9 x 10(-40) cm(6) s(-1), C(Yb-Yb) = 42 x 10(-40) cm(6) s(-1) and C(Tm-Tm) = 43 x 10(-40) cm(6) s(-1), respectively.

Testing covariance stationarity

In this paper, we show that the widely used stationarity tests such as the KPSS test have power close to size in the presence of time-varying unconditional variance. We propose a new test as a complement of the existing tests. Monte Carlo experiments show that the proposed test possesses the following characteristics: (i) In the presence of unit root or a structural change in the mean, the proposed test is as powerful as the KPSS and other tests; (ii) In the presence a changing variance, the traditional tests perform badly whereas the proposed test has high power comparing to the existing tests; (iii) The proposed test has the same size as traditional stationarity tests under the null hypothesis of stationarity. An application to daily observations of return on US Dollar/Euro exchange rate reveals the existence of instability in the unconditional variance when the entire sample is considered, but stability is found in subsamples.

Ensaios em finanças públicas municipais

Esta tese busca discutir problemas relacionados às finanças públicas municipais no país. Um aprimoramento de nosso federalismo fiscal, com maior descentralização de recursos e implementação de regras de transferências federativas com maiores incentivos pró-eficiência – aumentando os incentivos pró-obtenção de melhores resultados sociais por parte das unidades subfederadas – pode se revelar instrumento importante na melhoria de nossas condições sociais. Para isto, dividiu-se esta tese em quatro partes distintas. No primeiro capítulo, discutem-se os impactos das receitas petrolíferas sobre as finanças públicas municipais. As mudanças legais ocorridas no país na década de noventa do último século - relativas ao setor petrolífero - levaram a crescente (e concentrada) transferência de recursos do setor para os estados e municípios brasileiros. A forte sensação que estes estariam sendo desperdiçados de alguma forma vem suscitando discussões sobre a necessidade de se reformular sua distribuição. As recentes descobertas de megacampos petrolíferos no pré-sal do litoral brasileiro somente intensificaram este processo. Buscou-se identificar os efeitos destas transferências sobre as variáveis fiscais municipais no país. Detectou-se que não ocorreu substituição tributária, ou seja, estes recursos não diminuíram o esforço arrecadatório dos municípios. Em compensação, tanto os recursos cuja distribuição é bem mais concentrada (referentes aos royalties excedentes) quanto aqueles mais bem-distribuídos (referentes aos royalties originais) levaram as prefeituras a aumentar seus gastos correntes (piorando sua composição do ponto de vista social) e diminuir seus investimentos. O contrário parece ocorrer com os recursos cuja distribuição se dá de forma intermediária (os royalties referentes às participações especiais). No segundo capítulo, discute-se o impacto das receitas petrolíferas sobre a proficiência dos alunos até a quarta série primária das escolas públicas municipais. As receitas petrolíferas – agregadas ou não – não se mostraram estatisticamente significativas na explicação do desempenho observado pelos alunos da quarta série primária das escolas municipais em português ou matemática. Este resultado, entretanto, deve ser olhado com cautela, uma vez que não é trivial identificar como (e o tempo necessário) estes efeitos seriam gerados. Entretanto, diferentes fontes de receitas municipais impactariam de forma diferenciada as proficiências observadas nos testes de português e matemática, explicitando a necessidade de se entender melhor estas diferenças para se desenhar mecanismo mais eficiente de repasse de recursos constitucionais aos municípios. No terceiro capítulo, estudam-se os impactos das emancipações municipais ocorridas na década de 90 sobre o bem-estar das populações locais. Devido à Constituição Federal de 1988, o número de municípios no Brasil multiplicou-se fortemente na década de 90 do último século. Mais de mil municípios foram criados em todo o país, fazendo seu número ultrapassar a casa dos 5.500. Este processo tem sido interpretado de forma bastante negativa. Baseado em evidências anedóticas, se pressupõe que os atores políticos locais o utilizaram para se apropriar de maior parcela dos recursos transferidos de outros níveis governamentais. Entretanto, nenhum esforço mais sistemático foi realizado buscando calcular, de maneira efetiva, os resultados sociais líquidos deste processo. É isto que se busca fazer aqui, utilizando dados sobre os municípios mineiros - cujo número passa de 723 em 1991 para 853 em 2000. Foram detectados impactos positivos relacionados a diversas variáveis educacionais e de saúde. Ao mesmo tempo, o contrário ocorreu com os indicadores de pobreza e indigência. Este resultado mostra que o movimento observado de emancipação municipal talvez tenha sido bastante benéfico, sinalizando para a existência de mercados políticos eficientes nestas localidades, o que indicaria a necessidade de se manter uma maior autonomia local relativa a processos de emancipação de distritos. Por fim, no quarto capítulo, analisam-se os impactos de lei (implementada em Minas Gerais) que buscou aumentar os incentivos pró-eficiência das prefeituras municipais a partir das transferências federativas. Buscando melhorar as condições de vida dos municípios mineiros, o governo estadual mineiro instituiu, em 1995, a lei 12.040, conhecida como Lei Robin Hood. Esta permitia que 25% dos recursos de ICMS a serem distribuídos aos municípios mineiros se dessem sobre resultados observáveis em diversas áreas tais como saúde, educação, conservação ecológica, entre outras. Ou seja, esta instituía, em relação a estas transferências, um contrato de alto poder com os municípios relacionados às políticas públicas implementadas. O estudo destas transferências (relativas à educação e saúde) mostrou resultados dúbios. Resultados positivos relativos à educação e à saúde parecem ocorrer em todo o estado, mas os incentivos dados poderiam ser bem maiores - faz-se necessário um refinamento das regras da partilha destas transferências. Dada a relativa escassez deste tipo de contrato em transferências federativas, seja no Brasil, seja no restante do mundo, este resultado aponta a necessidade de utilização de instrumentos de maior poder nas relações federativas, buscando incrementar as condições de vida locais.