Factor analysis based VTS discriminative adaptive training

Vector Taylor Series (VTS) model based compensation is a powerful approach for noise robust speech recognition. An important extension to this approach is VTS adaptive training (VAT), which allows canonical models to be estimated on diverse noise-degraded training data. These canonical model can be estimated using EM-based approaches, allowing simple extensions to discriminative VAT (DVAT). However to ensure a diagonal corrupted speech covariance matrix the Jacobian (loading matrix) relating the noise and clean speech is diagonalised. In this work an approach for yielding optimal diagonal loading matrices based on minimising the expected KL-divergence between the diagonal loading matrix and "correct" distributions is proposed. The performance of DVAT using the standard and optimal diagonalisation was evaluated on both in-car collected data and the Aurora4 task. © 2012 IEEE.

Series-parallel six-element synthesis of biquadratic impedances

The object of this paper is to give a complete treatment of the realizability of positive-real biquadratic impedance functions by six-element series-parallel networks comprising resistors, capacitors, and inductors. This question was studied but not fully resolved in the classical electrical circuit literature. Renewed interest in this question arises in the synthesis of passive mechanical impedances. Recent work by the authors has introduced the concept of a regular positive-real functions. It was shown that five-element networks are capable of realizing all regular and some (but not all) nonregular biquadratic positive-real functions. Accordingly, the focus of this paper is on the realizability of nonregular biquadratics. It will be shown that the only six-element series-parallel networks which are capable of realizing nonregular biquadratic impedances are those with three reactive elements or four reactive elements. We identify a set of networks that can realize all the nonregular biquadratic functions for each of the two cases. The realizability conditions for the networks are expressed in terms of a canonical form for biquadratics. The nonregular realizable region for each of the networks is explicitly characterized. © 2004-2012 IEEE.

Numerical analysis of AC loss reduction in HTS superconducting coils using magnetic materials to divert flux

In this paper, the use of magnetic materials to divert flux in high-temperature superconductor superconducting coils and reduce transport ac loss is investigated. This particular technique is preferred over other techniques, such as striation, Roebel transposition, and twisted wires because it does not require modification to the conductor itself, which can be detrimental to the properties of the superconductor. The technique can also be implemented for existing coils. The analysis is carried out using a coil model based on the H formulation and implemented in comsol multiphysics. Both weakly and strongly magnetic materials are investigated, and it is shown that the use of such materials can divert flux and achieve a reduction in transport ac loss, which, in some cases, is quite significant. This analysis acts to provide a foundation for further optimization and experimental work in the future. © 2011 IEEE.

3D modeling of high-T c superconductors by finite element software

A three-dimensional (3D) numerical model is proposed to solve the electromagnetic problems involving transport current and background field of a high-T c superconducting (HTS) system. The model is characterized by the E-J power law and H-formulation, and is successfully implemented using finite element software. We first discuss the model in detail, including the mesh methods, boundary conditions and computing time. To validate the 3D model, we calculate the ac loss and trapped field solution for a bulk material and compare the results with the previously verified 2D solutions and an analytical solution. We then apply our model to test some typical problems such as superconducting bulk array and twisted conductors, which cannot be tackled by the 2D models. The new 3D model could be a powerful tool for researchers and engineers to investigate problems with a greater level of complicity.

DNS of EGR-type turbulent flame in MILD condition

Three-dimensional direct numerical simulation (DNS) of exhaust gas recirculation (EGR)-type turbulent combustion operated in moderate and intense low-oxygen dilution (MILD) condition has been carried out to study the flame structure and flame interaction. In order to achieve adequate EGR-type initial/inlet mixture fields, partially premixed mixture fields which are correlated with the turbulence are carefully preprocessed. The chemical kinetics is modelled using a skeletal mechanism for methane-air combustion. The results suggest that the flame fronts have thin flame structure and the direct link between the mean reaction rate and scalar dissipation rate remains valid in the EGR-type combustion with MILD condition. However, the commonly used canonical flamelet is not fully representative for MILD combustion. During the flame-flame interactions, the heat release rate increases higher than the maximum laminar flame value, while the gradient of progress variable becomes smaller than laminar value. It is also proposed that the reaction rate and the scalar gradient can be used as a marker for the flame interaction. © 2012 The Combustion Institute. Published by Elsevier Inc. All rights reserved.

Fabrication, densification, and replica molding of 3D carbon nanotube microstructures

The introduction of new materials and processes to microfabrication has, in large part, enabled many important advances in microsystems, labon- a-chip devices, and their applications. In particular, capabilities for cost-effective fabrication of polymer microstructures were transformed by the advent of soft lithography and other micromolding techniques 1,2, and this led a revolution in applications of microfabrication to biomedical engineering and biology. Nevertheless, it remains challenging to fabricate microstructures with well-defined nanoscale surface textures, and to fabricate arbitrary 3D shapes at the micro-scale. Robustness of master molds and maintenance of shape integrity is especially important to achieve high fidelity replication of complex structures and preserving their nanoscale surface texture. The combination of hierarchical textures, and heterogeneous shapes, is a profound challenge to existing microfabrication methods that largely rely upon top-down etching using fixed mask templates. On the other hand, the bottom-up synthesis of nanostructures such as nanotubes and nanowires can offer new capabilities to microfabrication, in particular by taking advantage of the collective self-organization of nanostructures, and local control of their growth behavior with respect to microfabricated patterns. Our goal is to introduce vertically aligned carbon nanotubes (CNTs), which we refer to as CNT "forests", as a new microfabrication material. We present details of a suite of related methods recently developed by our group: fabrication of CNT forest microstructures by thermal CVD from lithographically patterned catalyst thin films; self-directed elastocapillary densification of CNT microstructures; and replica molding of polymer microstructures using CNT composite master molds. In particular, our work shows that self-directed capillary densification ("capillary forming"), which is performed by condensation of a solvent onto the substrate with CNT microstructures, significantly increases the packing density of CNTs. This process enables directed transformation of vertical CNT microstructures into straight, inclined, and twisted shapes, which have robust mechanical properties exceeding those of typical microfabrication polymers. This in turn enables formation of nanocomposite CNT master molds by capillary-driven infiltration of polymers. The replica structures exhibit the anisotropic nanoscale texture of the aligned CNTs, and can have walls with sub-micron thickness and aspect ratios exceeding 50:1. Integration of CNT microstructures in fabrication offers further opportunity to exploit the electrical and thermal properties of CNTs, and diverse capabilities for chemical and biochemical functionalization 3. © 2012 Journal of Visualized Experiments.

Riemannian geometry of Grassmann manifolds with a view on algorithmic computation

We give simple formulas for the canonical metric, gradient, Lie derivative, Riemannian connection, parallel translation, geodesics and distance on the Grassmann manifold of p-planes in ℝn. In these formulas, p-planes are represented as the column space of n × p matrices. The Newton method on abstract Riemannian manifolds proposed by Smith is made explicit on the Grassmann manifold. Two applications - computing an invariant subspace of a matrix and the mean of subspaces - are worked out.

PCR-DGGE Fingerprinting Analysis of Plankton Communities and Its Relationship to Lake Trophic Status

Plankton communities in eight lakes of different trophic status near Yangtze, China were characterized by using denatured gradient gel electrophoresis (DGGE). Various water quality parameters were also measured at each collection site. Following extraction of DNA from plankton communities, 16S rRNA and 18S rRNA genes were amplified with specific primers for prokaryotes and eukaryotes, respectively; DNA profiles were developed by DGGE. The plankton community of each lake had its own distinct DNA profile. The total number of bands identified at 34 sampling stations ranged from 37 to 111. Both prokaryotes and eukaryotes displayed complex fingerprints composed of a large number of bands: 16 to 59 bands were obtained with the prokaryotic primer set; 21 to 52 bands for the eukaryotic primer set. The DGGE-patterns were analyzed in relation to water quality parameters by canonical correspondence analysis (CCA). Temperature, pH, alkalinity, and the concentration of COD, TP and TN were strongly correlated with the DGGE patterns. The parameters that demonstrated a strong correlation to the DGGE fingerprints of the plankton community differed among lakes, suggesting that differences in the DGGE fingerprints were due mainly to lake trophic status. Results of the present study suggest that PCR-DGGE fingerprinting is an effective and precise method of identifying changes to plankton community composition, and therefore could be a useful ecological tool for monitoring the response of aquatic ecosystems to environmental perturbations.

Plankton Community Succession in Artificial Systems Subjected to Cyanobacterial Blooms Removal using Chitosan-Modified Soils

Using artificial systems to simulate natural lake environments with cyanobacterial blooms, we investigated plankton community succession by polymerase chain reaction-denaturing gradient gel electrophoresis (PCR-DGGE) fingerprinting and morphological method. With this approach, we explored potential ecological effects of a newly developed cyanobacterial blooms removal method using chitosan-modified soils. Results of PCR-DGGE and morphological identification showed that plankton communities in the four test systems were nearly identical at the beginning of the experiment. After applying the newly developed and standard removal methods, there was a shift in community composition, but neither chemical conditions nor plankton succession were significantly affected by the cyanobacteria removal process. The planted Vallisneria natans successfully recovered after cyanobacteria removal, whereas that in the box without removal process did not. Additionally, canonical correspondence analysis indicated that other than for zooplankton abundance, total phosphorus was the most important environmental predictor of planktonic composition. The present study and others suggest that dealing with cyanobacteria removal using chitosan-modified soils can play an important role in controlling cyanobacterial blooms in eutrophicated freshwater systems.

Spatial distribution of benthic algae in the Gangqu River, Shangrila, China

This study consisted of sampling benthic algae at 32 sites in the Gangqu River, an important upstream tributary of the Yangtze River. Our aims were to characterize the benthic algae communities and relationships with environmental variables. Among the 162 taxa observed, Achnanthes linearis and Achnanthes lanceolata var. elliptica were the dominant species (17.10% and 14.30% of the total relative abundance, respectively). Major gradients and principal patterns of variation within the environmental variables were detected by principal component analysis (PCA). Then non-metric multidimensional scaling (NMS) divided all the sites into three groups, which were validated by multi-response permutation procedures (MRPP). Canonical correspondence analysis (CCA) indicated that three environmental variables (TN, TDS, and TP) significantly affected the distribution of benthic algae. Weighted averaging regression and cross-calibration produced strong models for predicting TN and TDS concentration, which enabled selection of algae taxa as potentially sensitive indicators of certain TN and TDS levels: for TN, Achnanthes lanceolata, Achnanthes lanceolata var. elliptica, and Cymbella ventricosa var. semicircularis; for TDS, Cocconeis placentula, Cymbella alpina var. minuta, and Fragilaria virescens. The present study represents an early step in establishing baseline conditions. Further monitoring is suggested to gain a better understanding of this region.

Saprolegnia brachydanis, A New Oomycete Isolated from Zebra Fish

Saprolegnia brachydanis is described from zebra fish (Brachydanio rerio) in Wuhan, Hubei Province, China. The species is illustrated and compared with other species of the genus. The distinctive characteristics of S. brachydanis are the production of glomerulate oogonia wrapped around by predominantly monoclinous antheridia which can be up to eight in one oogonium. The oogonial stalks are short, straight, or curved and the antheridia, twisted, can enwind one or more oogonia. The oospores cannot mature or easily abort. Morphological features of the oomycete and the ITS sequence of its rDNA as well as the comparison with related species are discussed in this article.

Randomized nonlinear component analysis

Copyright © (2014) by the International Machine Learning Society (IMLS) All rights reserved. Classical methods such as Principal Component Analysis (PCA) and Canonical Correlation Analysis (CCA) are ubiquitous in statistics. However, these techniques are only able to reveal linear re-lationships in data. Although nonlinear variants of PCA and CCA have been proposed, these are computationally prohibitive in the large scale. In a separate strand of recent research, randomized methods have been proposed to construct features that help reveal nonlinear patterns in data. For basic tasks such as regression or classification, random features exhibit little or no loss in performance, while achieving drastic savings in computational requirements. In this paper we leverage randomness to design scalable new variants of nonlinear PCA and CCA; our ideas extend to key multivariate analysis tools such as spectral clustering or LDA. We demonstrate our algorithms through experiments on real- world data, on which we compare against the state-of-the-art. A simple R implementation of the presented algorithms is provided.

Response of macrozoobenthos communities to ecological engineering remediation in a hypertrophic urban lake

The relationship of macrozoobenthos communities with some environmental variables, and their response to the ongoing restoration measures were studied in a small hypertrophic urban lake near the Yangzte River, China. Twenty taxa including 9 oligochaetes, 7 insects, 2 mollusks and two other animals were found during March 2005 to May 2006. The reappearance of some indigenous macrozoobenthos species showed that the ecological engineering remediation carried out was helpful for the recovery of the macrozoobenthos communities. Through canonical correspondence analysis (CCA), it was detected that temperature (T), conductivity (COND), Secchi depth/deep (SD/Deep), TN, total suspended solids (SS) and chemical oxygen demand (CODcr) were significant environmental factors that influenced the pattern of macrozoobenthos. Limnodrilus hoffmeisteri, Tanypus sp. and Alocinma longicornis could be used as potential bio-indicators in monitoring the development of ecological restoration.

Plankton community composition in the Three Gorges Reservoir Region revealed by PCR-DGGE and its relationships with environmental factors

To explore the relationships between community composition and the environment in a reservoir ecosystem, plankton communities from the Three Gorges Reservoir Region were studied by PCR-denaturing gradient gel electrophoresis fingerprinting. Bacterial and eukaryotic operational taxonomic units (OTUs), generated by DGGE analysis of the PCR-amplified 16S and 18S rRNA genes, were used as surrogates for the dominant "biodiversity units". OTU composition among the sites was heterogeneous; 46.7% of the total bacteria] OTUs (45) and 64.1% of the eukaryotic OTUs (39) were identified in less than half of the sampling sites. Unweighted pair group method with arithmetic averages (UPGMA) clustering of the OTUs suggested that the plankton communities in the Xiangxi Rive sites were not always significantly different from those from the Yangtze River sites, despite clear differences in their environmental characterizations. Canonical correspondence analysis (CCA) was applied to further investigate the relationships between OTU composition and the environmental factors. The first two CCA ordination axes suggested that the bacterial community composition was primarily correlated with the variables of NO3--N, dissolved oxygen (DO), and SiO32--Si, whereas, the eukaryotic community was mainly correlated with the concentrations of DO, PO43--P, and SiO32--Si.

Digital radio distribution system over multiple cabling types

This paper demonstrates a novel digital radio distribution system able to transmit not only over optical fibres and coaxial cables but also over twisted pair cables. The digitised RF signal is compressed for maximum transmission efficiency in a way that allows for integral self-learning algorithms to be introduced for multi-service applications. © 2013 IEEE.