Construction of tunable radial basis function networks using orthogonal forward selection

An orthogonal forward selection (OFS) algorithm based on leave-one-out (LOO) criteria is proposed for the construction of radial basis function (RBF) networks with tunable nodes. Each stage of the construction process determines an RBF node, namely, its center vector and diagonal covariance matrix, by minimizing the LOO statistics. For regression application, the LOO criterion is chosen to be the LOO mean-square error, while the LOO misclassification rate is adopted in two-class classification application. This OFS-LOO algorithm is computationally efficient, and it is capable of constructing parsimonious RBF networks that generalize well. Moreover, the proposed algorithm is fully automatic, and the user does not need to specify a termination criterion for the construction process. The effectiveness of the proposed RBF network construction procedure is demonstrated using examples taken from both regression and classification applications.

Probability density estimation with tunable kernels using orthogonal forward regression

A generalized or tunable-kernel model is proposed for probability density function estimation based on an orthogonal forward regression procedure. Each stage of the density estimation process determines a tunable kernel, namely, its center vector and diagonal covariance matrix, by minimizing a leave-one-out test criterion. The kernel mixing weights of the constructed sparse density estimate are finally updated using the multiplicative nonnegative quadratic programming algorithm to ensure the nonnegative and unity constraints, and this weight-updating process additionally has the desired ability to further reduce the model size. The proposed tunable-kernel model has advantages, in terms of model generalization capability and model sparsity, over the standard fixed-kernel model that restricts kernel centers to the training data points and employs a single common kernel variance for every kernel. On the other hand, it does not optimize all the model parameters together and thus avoids the problems of high-dimensional ill-conditioned nonlinear optimization associated with the conventional finite mixture model. Several examples are included to demonstrate the ability of the proposed novel tunable-kernel model to effectively construct a very compact density estimate accurately.

Adaptive modelling with tunable RBF network using multi-innovation RLS algorithm assisted by swarm intelligence

In this paper, we propose a new on-line learning algorithm for the non-linear system identification: the swarm intelligence aided multi-innovation recursive least squares (SI-MRLS) algorithm. The SI-MRLS algorithm applies the particle swarm optimization (PSO) to construct a flexible radial basis function (RBF) model so that both the model structure and output weights can be adapted. By replacing an insignificant RBF node with a new one based on the increment of error variance criterion at every iteration, the model remains at a limited size. The multi-innovation RLS algorithm is used to update the RBF output weights which are known to have better accuracy than the classic RLS. The proposed method can produces a parsimonious model with good performance. Simulation result are also shown to verify the SI-MRLS algorithm.

pH-tunable hydrogelators for water purification: structural optimisation and evaluation

A focused library of potential hydrogelators each containing two substituted aromatic residues separated by a urea or thiourea linkage have been synthesised and characterized. Six of these novel compounds are highly efficient hydrogelators, forming gels in aqueous solution at low concentrations (0.03–0.60 wt %). Gels were formed through a pH switching methodology, by acidification of a basic solution (pH 14 to ≈4) either by addition of HCl or via the slow hydrolysis of glucono-δ-lactone. Frequently, gelation was accompanied by a dramatic switch in the absorption spectra of the gelators, resulting in a significant change in colour, typically from a vibrant orange to pale yellow. Each of the gels was capable of sequestering significant quantities of the aromatic cationic dye, methylene blue, from aqueous solution (up to 1.02 g of dye per gram of dry gelator). Cryo-transmission electron microscopy of two of the gels revealed an extensive network of high aspect ratio fibers. The structure of the fibers altered dramatically upon addition of 20 wt % of the dye, resulting in aggregation and significant shortening of the fibrils. This study demonstrates the feasibility for these novel gels finding application as inexpensive and effective water purification platforms.

Adaptive noise cancellation with fast tunable RBF network

This paper describes a novel adaptive noise cancellation system with fast tunable radial basis function (RBF). The weight coefficients of the RBF network are adapted by the multi-innovation recursive least square (MRLS) algorithm. If the RBF network performs poorly despite of the weight adaptation, an insignificant node with little contribution to the overall performance is replaced with a new node without changing the model size. Otherwise, the RBF network structure remains unchanged and only the weight vector is adapted. The simulation results show that the proposed approach can well cancel the noise in both stationary and nonstationary ANC systems.

Online modeling with tunable RBF network

In this paper, we propose a novel online modeling algorithm for nonlinear and nonstationary systems using a radial basis function (RBF) neural network with a fixed number of hidden nodes. Each of the RBF basis functions has a tunable center vector and an adjustable diagonal covariance matrix. A multi-innovation recursive least square (MRLS) algorithm is applied to update the weights of RBF online, while the modeling performance is monitored. When the modeling residual of the RBF network becomes large in spite of the weight adaptation, a node identified as insignificant is replaced with a new node, for which the tunable center vector and diagonal covariance matrix are optimized using the quantum particle swarm optimization (QPSO) algorithm. The major contribution is to combine the MRLS weight adaptation and QPSO node structure optimization in an innovative way so that it can track well the local characteristic in the nonstationary system with a very sparse model. Simulation results show that the proposed algorithm has significantly better performance than existing approaches.

A fast adaptive tunable RBF network for nonstationary systems

This paper describes a novel on-line learning approach for radial basis function (RBF) neural network. Based on an RBF network with individually tunable nodes and a fixed small model size, the weight vector is adjusted using the multi-innovation recursive least square algorithm on-line. When the residual error of the RBF network becomes large despite of the weight adaptation, an insignificant node with little contribution to the overall system is replaced by a new node. Structural parameters of the new node are optimized by proposed fast algorithms in order to significantly improve the modeling performance. The proposed scheme describes a novel, flexible, and fast way for on-line system identification problems. Simulation results show that the proposed approach can significantly outperform existing ones for nonstationary systems in particular.

Sparse density estimator with tunable kernels

A new sparse kernel density estimator with tunable kernels is introduced within a forward constrained regression framework whereby the nonnegative and summing-to-unity constraints of the mixing weights can easily be satisfied. Based on the minimum integrated square error criterion, a recursive algorithm is developed to select significant kernels one at time, and the kernel width of the selected kernel is then tuned using the gradient descent algorithm. Numerical examples are employed to demonstrate that the proposed approach is effective in constructing very sparse kernel density estimators with competitive accuracy to existing kernel density estimators.

Photoluminescence behavior of Eu(3+) ion doped into gamma- and alpha-alumina systems prepared by combustion, ceramic and Pechini methods

Al(2)O(3):Eu(3+)(1%) samples were prepared by combustion, ceramic, and Pechini methods annealed from 400 to 1400 degrees C. XRD patterns indicate that samples heated up to 1000 degrees C present disordered character of activated alumina (gamma-Al(2)O(3)). However, alpha-Al(2)O(3) phase showed high crystallinity and thermostability at 1200-1400 degrees C. The sample characterizations were also carried out by means of infrared spectroscopy (IR), scanning electron microscopy (SEM) and specific surface areas analysis (BET method). Excitation spectra of Al(2)O(3):Eu(3+) samples present broaden bands attributed to defects of Al(2)O(3) matrices and to LMCT state of O -> Eu(3+), however, the narrow bands are assigned to (7)F(0) -> (5)D(J),(5)H(J) and (5)L(J) transitions of Eu(3+) ion. Emission spectra of samples calcined up to 1000 degrees C show broaden bands for (5)D(0) -> (7)F(J) transitions of Eu(3+) ion suggesting that the rare earth ion is in different symmetry sites showed by inhomogeneous line broadening of bands, confirming the predominance of the gamma-alumina phase. For all samples heated from 1200 to 1400 degrees C the spectra exhibit narrow (5)D(0) -> (7)F(J) transitions of Eu(3+) ion indicating the conversion of gamma to alpha-Al(2)O(3) phases, a high intensity narrow peak around 695 nm assigned to R lines of Cr(3+) ion is shown. Al(2)O(3):Eu(3+) heated up to 1100 degrees C presents an increase in the Omega(2) intensity parameter with the increase of temperatures enhancing the covalent character of metal-donor interaction. The disordered structural systems present the highest values of emission quantum efficiencies (eta). CIE coordinates of Al(2)O(3):Eu(3+) are also discussed. (C) 2007 Elsevier Inc. All rights reserved.

(85)Rb Bose-Einstein condensate with tunable interaction: A quantum many body approach

We present a quantum many body approach with van der Waal type of interaction to achieve (85)Rb Bose-Einstein condensate with tunable interaction which has been produced by magnetic field induced Feshbach resonance in the JILA experiment. (C) 2008 Elsevier B.V. All rights reserved.

Characterization, photoluminescence, thermally stimulated luminescence and electron spin resonance studies of Eu(3+) doped LaAlO(3) phosphor

Europium-doped lanthanum aluminate (LaAlO(3)) powder was prepared by using a combustion method. The crystallization, surface morphology, specific surface area and luminescence properties of the samples have been investigated. Photoluminescence studies of Eu doped LaAlO(3) showed orange-reddish emission due to Eu(3+) ions. LaAlO(3):Eu(3+) exhibits one thermally stimulated luminescence (TSL) peak around 400 degrees C. Room temperature electron spin resonance spectrum of irradiated phosphor appears to be a superposition of two centres. One of them (centre I) with principal g-value 2.017 is identified as an O(-) centre while centre II with an isotropic g-value 2.011 is assigned to an F(+) centre (singly ionized oxygen vacancy). An additional defect centre observed during thermal annealing around 300 degrees C grows with the annealing temperature. This centre (assigned to F(+) centre) originates from an F-centre (oxygen vacancy with two electrons) and the F-centre along with the associated F(+) centre appear to correlate with the observed TSL peak in LaAlO(3):Eu(3+) phosphor. The activation energy for this peak has been determined to be 1.54 eV from TSL data. (C) 2010 Elsevier Masson SAS. All rights reserved.

A new processing method of CaZn(2)(OH)(6)center dot 2H(2)O powders: Photoluminescence and growth mechanism

In this communication, we report on the formation of calcium hexahydroxodizincate dehydrate, CaZn(2)(OH)(6)center dot 2H(2)O (CZO) powders under microwave-hydrothermal (MH) conditions. These powders were analyzed by X-ray diffraction (XRD), Field-emission gum scanning electron microscopy (FEG-SEM), ultraviolet-visible (UV-vis) absorption spectroscopy and photoluminescence (PL) measurements. XRD patterns confirmed that the pure CZO phase was obtained after MH processing performed at 130 degrees C for 2 h. FEG-SEM micrographs indicated that the morphological modifications as well as the growth of CZO microparticles are governed by Ostwald-ripening and coalescence mechanisms. UV-vis spectra showed that this material have an indirect optical band gap. The pure CZO powders exhibited an yellow PL emission when excited by 350 nm wavelength at room temperature. (C) 2009 Elsevier Masson SAS. All rights reserved.

Optical and dielectric relaxor behaviour of Ba(Zr(0.25)Ti(0.75))O(3) ceramic explained by means of distorted clusters

In this work, Ba(Zr(0.25)Ti(0.75))O(3) ceramic was prepared by solid-state reaction. This material was characterized by x-ray diffraction and Fourier transform Raman spectroscopy. The temperature dependent dielectric properties were investigated in the frequency range from 1 kHz to 1 MHz. The dielectric measurements indicated a diffuse phase transition. The broadening of the dielectric permittivity in the frequency range as well as its shifting at higher temperatures indicated a relaxor-like behaviour for this material. The diffusivity and the relaxation strength were estimated using the modified Curie-Weiss law. The optical properties were analysed by ultraviolet-visible (UV-vis) absorption spectroscopy and photoluminescence (PL) measurements at room temperature. The UV-vis spectrum indicated that the Ba(Zr(0.25)Ti(0.75))O(3) ceramic has an optical band gap of 2.98 eV. A blue PL emission was observed for this compound when excited with 350 nm wavelength. The polarity as well as the PL property of this material was attributed to the presence of polar [TiO(6)] distorted clusters into a globally cubic matrix.

Morphology and Blue Photoluminescence Emission of PbMoO(4) Processed in Conventional Hydrothermal

PbMoO(4) micro-octahedrons were prepared by the coprecipitation method at room temperature without the presence of surfactants and processed in a conventional hydrothermal at different temperatures (from 60 to 120 degrees C) for 10 min. These micro-octahedrons were structurally characterized by X-ray diffraction (XRD) and micro-Raman (MR) spectroscopy, and its morphology was investigated by field-emission gun scanning electron microscopy (FEG-SEM). The optical properties were analyzed by ultraviolet-visible (UV-vis) absorption spectroscopy and photoluminescence (PL) measurements. XRD patterns and MR spectra confirmed that the PbMoO(4) micro-octahedrons are characterized by a scheelite-type tetragonal structure. FEG-SEM micrographs points, out that these structures present a polydisperse particle size distribution in consequence of a predominant growth mechanism via aggregation of particles. In addition, it was observed that the hydrothermal conditions favored a spontaneous formation of micro-octahedrons interconnected along a common crystallographic orientation (oriented-attachment), resulting in self-organized structures. An intense blue PL emission at room temperature was observed in these micro-octahedrons when they were excited with a 350 nm wavelength. The origin of the PL emissions as well as its intensity variations are explained by means of a model based on both distorted [MoO(4)] and [PbO(8)] clusters into the lattice.

Intense violet-blue photoluminescence in BaZrO3 powders: A theoretical and experimental investigation of structural order-disorder

Intense violet-blue photoluminescence (PL) emission at room temperature was verified in BaZrO3 (BZO) powders with structural order-disorder. Ab-initio calculations, ultraviolet-visible absorption spectroscopy and PL were performed. Theoretical results showed that the local disorder in the network-formed Zr clusters present an important role in the formation of hole-electron pair. The experimental data and theoretical results are in agreement, indicating that the PL emission in BZO powders can be related to the structural order-disorder degree in the lattice. (C) 2008 Elsevier B.V. All rights reserved.