Pair-copula based mixture models and their application in clustering

Finite mixtures are often used to perform model based clustering of multivariate data sets. In real life applications, such data may exhibit complex nonlinear form of dependence among the variables. Also, the individual variables (margins) may follow different families of distributions. Most of the existing mixture models are unable to accommodate these two aspects of the data. This paper presents a finite mixture model that involves a pair-copula based construction of a multivariate distribution. Such a model de-couples the margins and the dependence structures. Hence, the margins can be modeled using different families. Again, many possible dependence structures can also be studied using different copulas. The resulting mixture model (called DVMM) is then capable of capturing a broad family of distributions including non-Gaussian models. Here we study DVMM in the context of clustering of multivariate data. We design an expectation maximization procedure for estimating the mixture parameters. We perform extensive experiments on the basis of a number of well-known data sets. A detailed evaluation of the clustering quality obtained by DVMM in comparison to other mixture models is presented. The experimental results show that the performance of DVMM is quite satisfactory.     

SWGMM: a semi-wrapped Gaussian mixture model for clustering of circular–linear data

Finite mixture models are widely used to perform model-based clustering of multivariate data sets. Most of the existing mixture models work with linear data; whereas, real-life applications may involve multivariate data having both circular and linear characteristics. No existing mixture models can accommodate such correlated circular–linear data. In this paper, we consider designing a mixture model for multivariate data having one circular variable. In order to construct a circular–linear joint distribution with proper inclusion of correlation terms, we use the semi-wrapped Gaussian distribution. Further, we construct a mixture model (termed SWGMM) of such joint distributions. This mixture model is capable of approximating the distribution of multi-modal circular–linear data. An unsupervised learning of the mixture parameters is proposed based on expectation maximization method. Clustering is performed using maximum a posteriori criterion. To evaluate the performance of SWGMM, we choose the task of color image segmentation in LCH space. We present comprehensive results and compare SWGMM with existing methods. Our study reveals that the proposed mixture model outperforms the other methods in most cases.     

Gate‐Controlled Energy Barrier at a Graphene/Molecular Semiconductor Junction

The formation of an energy‐barrier at a metal/molecular semiconductor junction is a universal phenomenon which limits the performance of many molecular semiconductor‐based electronic devices, from field‐effect transistors to light‐emitting diodes. In general, a specific metal/molecular semiconductor combination of materials leads to a fixed energy‐barrier. However, in this work, a graphene/C₆₀ vertical field‐effect transistor is presented in which control of the interfacial energy‐barrier is demonstrated, such that the junction switches from a highly rectifying diode at negative gate voltages to a highly conductive nonrectifying behavior at positive gate voltages and at room temperature. From the experimental data, an energy‐barrier modulation of up to 660 meV, a transconductance of up to five orders of magnitude, and a gate‐modulated photocurrent are extracted. The ability to tune the graphene/molecular semiconductor energy‐barrier provides a promising route toward novel, high performance molecular devices.     

An MLP-based texture segmentation method without selecting a feature set

A texture segmentation technique which employs a multilayer perceptron (MLP) and does not consider the selection of features is presented in this paper. Thus, users can avoid selection and computation of the feature set and hence real-time segmentation may be possible. The technique apparently works in a fashion similar to our visual system whereby we do not consciously compute any feature for texture discrimination. A detailed study has been made for the selection of the network size. A newly proposed variant of the back-propagation algorithm has been used for more efficient training of the network. An edge-preserving noise-smoothing approach has been proposed to remove noise from the segmented image.     

Symmetry analysis by computer

Approximate locations of axes of symmetry of a 2-dimensional region are detected on the basis of its border. The border, described in terms of certain directional codes, is treated as a regular polygon in a hierarchical manner where a lower level means a greater number of sides. At each level of the hierarchy, the best axis of symmetry is found which for a lower level gives a more accurate position of the unknown axis of symmetry than for a higher level. Along with an axis of symmetry, a certain error is found on the basis of which the degree of symmetry of a 2-dimensional region is defined. Programs are written in FORTRAN IV and are implemented on an EC-1033 computer.     

Incompleteness of bifurcation diagram in predicting the behaviour of Ćuk converter in discontinuous conduction mode

The bifurcation phenomena have been studied in current mode‐controlled ?uk converter considering continuous conduction as well as discontinuous conduction mode of operation. Some peculiarities have been observed in discontinuous conduction mode, which indicate some incompleteness of the bifurcation diagrams in providing details of the system behaviour. Copyright © 2007 John Wiley & Sons, Ltd.     

Design of filtering directional coupler with improved performance

This letter presents a filtering directional coupler (FDC) with enhanced coupling and high directivity simultaneously. The proposed FDC is composed of a pair of coupled lines instead transmission line of a directional coupler. This coupled lines resonator increases the design parameters by which even/odd mode phase velocity can be compensated to improve the directivity and coupling level. The coupling enhancement can be explained by analyzing the even mode and odd mode circuit of the proposed coupler. A prototype of the proposed coupler is designed which provides a high directivity of 44 dB for 6 dB coupling level at 1 GHz frequency. The proposed coupler is designed, fabricated, and tested.     

A triple‐layer dual‐bandpass frequency selective surface of third order response with equivalent circuit analysis

A multilayered cascaded and polarization‐dependent frequency selective surface (FSS) exhibiting dual bandpass frequency response is proposed in this article. The FSS is composed of two metal‐based square patch layers in the two ends and one aperture type layer in the middle, separated by two dielectric substrates. The FSS exhibits bandpass response of third order with two transmission poles in the 5‐6 GHz band and one pole at 2.5 GHz. The passbands are separated well enough with a transmission zero at 3.5 GHz leading to significant out‐of‐band rejection. The structure is ultrathin with the thickness on the order of 0.01λ₀ with respect to the lowest resonating frequency. It is shown with parametric studies how the poles can be tuned individually. Principle of operation of the FSS is explained with its equivalent circuit model. Transmission phase of the FSS varies linearly with frequency in the upper band. Simulation result is verified experimentally for the fabricated prototype.     

Quasi-elliptic filter characteristics of an asymmetric defected ground structure

An asymmetric defected ground structure (DGS) with respect to a microstrip line is proposed. Its unit cell consists of two square slots connected with a rectangular slot by meandered transverse slots. The simulated, measured and equivalent circuit responses have been presented. The unit DGS pattern under microstrip line produces wide rejection bandwidth and high sharpness factor. Its elliptic band-reject filter response is modelled by appropriate LC equivalent circuit. Two such DGS cells under a microstrip line produce 3-pole quasi-elliptic bandstop filter response, which shows deeper attenuation without appreciable change in pole frequencies in comparison to single cell structure. Unit DGS pattern under coupled gap microstrip lines produces band-accept filtering characteristics. The frequency characteristics show a quasi-elliptic highpass filtering characteristics with sharp lower transition knee. Accordingly, an LC equivalent circuit is proposed. A cascaded band-accept and band-reject filter is proposed for designing a wideband bandpass filter. The bandwidth and centre frequency of the bandpass filter can be varied by choosing different set of DGS dimensions.