Publication9

Particle filters provide a robust framework for non-linear and non-Gaussian estimation problems. In this paper, we present a method to incorporate dominant modulation-domain (Amplitude Modulation-Frequency Modulation) features into particle filter based template tracking. The dominant AM-FM features capture the local texture structure of an image. Tracking is performed based on correlations of these AM-FM features. We propose an adaptive method that weights these correlations based on the significance of the features for tracking. The performance of the tracker is demonstrated on a few standard test sequences. © 2008 IEEE.

In this paper, we propose a machine learning approach to rhetorical role identification from legal documents. In our approach, we annotate roles in sample documents with the help of legal experts and take them as training data. Conditional random field model has been trained with the data to perform rhetorical role identification with reinforcement of rich feature sets. The understanding of structure of a legal document and the application of mathematical model can brings out an effective summary in the final stage. Other important new findings in this work include that the training of a model for one sub-domain can be extended to another sub-domains with very limited augmentation of feature sets. Moreover, we can significantly improve extraction-based summarization results by modifying the ranking of sentences with the importance of specific roles.

Surface quality of the machined component plays a vital role in the functional characteristics and life of the product. The objective of this work is divided into two phases: 1 To demonstrate a systematic procedure of using Taguchi parameter design in process control to identify the optimum surface roughness produced with a particular combination of cutting parameters; 2 To show the effectof high speed cutting parameters on the surface integrity aspects of Inconel 718. Surface integrity characteristics were assessed using surface roughness, micro hardness, micro structure, X-ray diffraction and SEM study of the machined surfaces, chip surfaces and tool surfaces. The results indicated that the surface and subsurface changes/alterations such as microstructural alteration, surface damage and phase change occur during machining. And it is clearly observed that the quality of the manufactured surfaces such as reduction of work hardening could be improved by proper selection of process parameters. © 2008, Inderscience Publishers.

This paper presents the findings of a study of motorized two-wheeler emissions under lane restricted and lane-less flow conditions in heterogeneous traffic. Lane restriction is defined as the movement of particular categories of vehicles on specified lanes (dedicated lanes), whereas lane-less flow movements are conditions when vehicles are free to change lanes. This study was conducted on selected mid-block road sections of different lengths in Chennai city, India. In the process of test runs, tailpipe emissions (CO, HC and NO) were measured using a portable gas analyzer at short intervals of time. These data were then synthesized to evaluate the total emissions from the vehicles for each test run. Instantaneous speeds of the vehicles were measured using an optical sensor fixed to the wheel. Consumption of fuel by the test vehicles were also measured using a petrol reading apparatus. Based on the analysis of field data collected specifically for this study, it is concluded that lane restricted flow generally produced reduced levels of tailpipe emissions compared to lane-less conditions. The impact of various factors such as, number of lane changes, average speed and fuel consumed on emissions were also studied. In order to understand the relationships between emissions and various traffic factors, regression models were tested and built using SPSS software. Highly heterogeneous traffic conditions and the lane changing behavior of drivers in traffic conditions prevailing in urban areas in India offered unique research opportunities in this study. Copyright ASCE 2008.

To illustrate the concept of the inter-dependence between multicomponent chemical reactions and transport phenomena, a theoretical investigation of a halogen-cycle tungsten filament incandescent lamp was undertaken based on local thermochemical equilibrium (LTCE) coupled with Fickian and thermal (Soret) transport of species. Deposition of tungsten on the bulb wall leads to 'bulb blackening', resulting in reduction in the luminous efficiency of the lamp, a problem typically combated by addition of small amounts of halogen and oxygen. Utilizing temperature-dependent Soret diffusion factors, elemental fluxes were analyzed. Zero element flux (ZEF) maps were generated based on temperature-dependent Soret diffusion factors. The element solubilities and the filament-to-wall element segregation were found to be strongly dependent on the bulb wall temperature, oxygen fill pressure and Soret diffusion factors. Transport-affected shifts in element solubilities from pure LTCE condition were studied. Implications of this study for the modelling of nano-scale CVD processes are highlighted. © The Electrochemical Society.

In this paper, we obtain sharp norm estimates for the Bernardi integral transform of functions belonging to the class K(A,B), -1≤B < A≤1 which is a subclass of the well-known class of convex univalent functions. As a consequence, a number of open questions arise naturally, concerning J (A,B)-a subclass of the wellknown class of starlike univalent functions, and many other classes.

For solving linear ill-posed problems with noisy data regularization methods are required. We analyze a simplified regularization scheme in Hilbert scales for operator equations with nonnegative self-adjoint operators. By exploiting the operator monotonicity of certain functions, order-optimal error bounds are derived that characterize the accuracy of the regularized approximations. These error bounds have been obtained under general smoothness conditions. © 2008, Institute of Mathematics, NAS of Belarus. All rights reserved.

The mechanical properties of the structural layer play an important role in the design and optimization of microelectromechanical system structures. The pull-in measurement is a popular technique used to measure the mechanical properties of a material, but its success depends on the accurate measurement of the gap (g) between the beam and the ground plane and its uniformity. We propose a novel technique that does not require accurate knowledge of the value of g. In our proposed method, a large number of beams with different lengths (L) are to be fabricated simultaneously and the off-capacitance (Coff) in addition to pull-in voltage (Vpi) measured in the same setup. To get accurate results, the range of length of beams must be properly chosen. We show, with the help of simulations, that by using our method the material properties can be extracted very accurately even when the gap (g) is nonuniform. © 2008 Society of Photo-Optical Instrumentation Engineers.

Optical burst switching (OBS) is considered as a contending technology for the core of the Internet in future. However, due to lack of the buffers, losses occur due to contention among simultaneously arriving bursts at the core nodes. Contention losses do not necessarily indicate a situation of congestion in the network. Thus differentiation (classification) of losses is essential in many applications to avoid false identification of congestion. In this paper, we propose a loss classification technique for the OBS networks based on machine learning techniques. We devise a new measure to differentiate between congestion and contention losses, which is derived from the observed losses, called the number of bursts between failures (NBBF). We observe that the NBBF follows a Gaussian distribution with different parameters for contention and congestion losses. This feature is used in differentiation. We use both a supervised learning technique (hidden Markov model (HMM)) and an unsupervised learning technique (expectation maximization (EM) clustering) on the observed losses and classify them into a set of states (clusters) after which an algorithm differentiates between the congestion and contention losses. We also demonstrate the use of loss differentiation in improving the performance of Transport Control Protocol (TCP) over OBS networks. We modify congestion control mechanism of TCP suitably to arrive at two variants of TCP, HMM-TCP and EM-TCP. Their performance is compared with TCP NewReno, TCP SACK, and Burst TCP [1]. Simulation results demonstrate the effectiveness and accuracy of the loss classification technique in different network scenarios. © 2008, IEEE. All rights reserved.

Dielectric resonators are widely used as components of microwave resonant cavi- ties with high dielectric constant, high quality factor and low temperature dependent resonant frequency. Recent advances in the synthesis of materials find several methods of preparation that includes microwave processing. It is possible to use the microwave for the synthesis of materials in two ways. Once the calcination of the required material is performed with conventional fur- nace, the microwave assisted sintering can be used for more densification. The other way is to directly use the microwave assisted process for obtaining the densified pellets from the stoichio- metrically mixed starting reagents. In this paper, we report the synthesis of Sm(Zn0.5Ti 0.5)O3 and ZnO-TiO2 using conventional sintering, microwave assisted sintering and microwave assisted synthesis. Phase formation was observed with XRD and surface morphology was observed using AFM. Microwave dielectric properties of all these systems were obtained using Hakki-Coleman method and cavity resonance method. The paper analyzes the variation between the microwave dielectric property and surface morphology of the above two systems synthesized by the three different routes. Copyright © (2008) by the Electromagnetics Academy.

An extensive program of laboratory model tests was undertaken to study the effect of slope on p-y curves due to surcharge load in dry sand. The paper concerns the method developed in a series of laboratory model tests to experimentally determine p-y curves. Bending moment curves are differentiated by using curve fitting method of cubic polynomial function. The study includes effect of slope angle and relative density on bending moment, lateral soil resistance, lateral deflection and non-dimensional p-y curves. The non-dimensional p-y curves for piles on sloping ground under surcharge load are developed modifying API RP 2A (2000) method by including a Reduction Factor (R) using the experimental results. Copyright© 2005-2009 National Institute of Informatics.

DC to impulse electric field measurements are performed by combining an electro-optic effect with the second harmonic generation in KTiOPO4 crystals. The technique is extended to measure orthogonal components of the electric field. © 2008 Optical Society of America.

In the present work, an attempt was made to study numerically, using CFD, the effect of vane-shape on the flow-field and heat transfer characteristics of a disc brake rotor for different configurations and at different speeds. Initially, the CFD code used in this work was validated by experimental results obtained by conducting experiments on a test rotor using particle image velocimetry (PIV). Further, six types of rotor configurations viz., straight radial vane (SRV), tapered radial vane (TRV), modified tapered radial vane (MTRV), circular pillared (CP), diamond pillared (DP) and modified diamond pillared (MDP) were considered for the numerical analysis. Three of them were radial type and other three were of pillared type rotors. A rotor segment of 20° was considered for the numerical analysis due to rotational symmetry. Validation was done for SRV rotor, for which the experimental and predicted results were in good agreement. The maximum deviation of the numerical results was about 10% from the experimental results. It was found that among the six types of rotor configurations considered for the analysis, MTRV rotor allows more air flow through the rotor (4.43 g/s at 1800 rpm), which was 35% more than that of SRV rotor. In addition, it was found that the heat dissipation from the surfaces of MTRV rotor was the highest i.e., about 20% more than SRV rotor and 9% more than MDP rotor. Among all the rotor configurations considered, the heat transfer from the passage surfaces was symmetrical in pillared rotors. Hence, for modern high-speed vehicles, ventilated pillared rotors may be more appropriate. Copyright © 2008 SAE International.

In this paper, we have proposed an analytical model that computes the blocking probability and the channel utilisation for a given load on the optical networks with wavelength conversion at the nodes. The computation has been carried out for fixed routing with uniform traffic distribution (UTD). When any link on a route blocks a call, the dropped call reduces the offered load on all other links invovled in the route. The overall blocking probability of the network is estimated using Erlang fixed point approximation with the reduced load on the links. The expression has been validated by comparing the results with that of simulation for a number of standard networks including 14 node NSFNET, 20 node ARPANET and 20 node INDIANET. We show that the analytical method performs well in the desired range of blocking probabilities and is applicable to any network topology. © 2008 by the IETE.

This paper presents an alternative numerical tool to model the one—and two—dimensional contaminant transport through saturated porous media using a meshfree method called Element Free Galerkin Method (EFGM). In EFGM, an approximate solution is constructed entirely in terms of a set of nodes and no characterization of the interrelationship of the nodes is needed. The advection-dispersion equation with sorption is considered to illustrate the applicability of EFGM. The unknown concentration values are approximated by using moving least-square approximants, which are constructed using a weight function, a linear basis function and a set of non-constant coefficients. The weak form of the governing equation is formulated and Lagrange multiplier method is used for enforcing the essential boundary conditions. Three numerical examples are presented and the results are compared with those obtained from the analytical and finite element results. © 2008 Taylor & Francis Group, LLC.

With ever-increasing demands on capacity, quality of service, speed, and reliability, current Internet systems are under strain and under review. Combining contributions from experts in the field, this book captures the most recent and innovative designs, architectures, protocols, and mechanisms that will enable researchers to successfully build the next-generation Internet. A broad perspective is provided, with topics including innovations at the physical/transmission layer in wired and wireless media, as well as the support for new switching and routing paradigms at the device and sub-system layer. The proposed alternatives to TCP and UDP at the data transport layer for emerging environments are also covered, as are the novel models and theoretical foundations proposed for understanding network complexity. Finally, new approaches for pricing and network economics are discussed, making this ideal for students, researchers, and practitioners who need to know about designing, constructing, and operating the next-generation Internet.

Cavitation erosion is predominant in pipelines for liquid transportation, causing damage to pipe wall, impeller and their accessories. The present study is focused on development of cavitation -wear resistant nano-ceramic particle-reinforced polymer matrix material; and on study of its feasibility to be used as lining material in hydraulic transportation. The polymer/nano composite is fabricated using power ultrasound in all three process steps: synthesis of nano-dimensional particles of white fused alumina (WFA) from micron size particles, optimized blending and finally reinforcement into poly methyl methacrylate (PMMA) matrix. The effect of ultrasonic parameters on nanocomposite/ virgin polymers (like polyethylene and polypropylene) is studied by measuring mass loss of the materials and suspension turbidity during exposure time. At low frequency (20-60 kHz), cavitation intensity is predominant; this effect is utilized for fabricating sub-micron particles, and for performing accelerated cavitation erosion tests. At high frequency, acoustic streaming is predominant; this effect is utilized for blending and reinforcing of the nano ceramic particles into polymer matrix. The size and quantity of the particles generated by cavitation erosion was analyzed by Laser Particle Size Analyzer (20 nm-1400 micron range). The nano-composite coupons were analyzed before and after the ultrasonic erosion test using SEM. It is concluded that lowfrequency sonication is a viable option for cavitaton erosion testing of ceramic/polymer composites.

Severe Plastic Deformation (SPD) processes like Equal Channel Angular Pressing (ECAP) are currently being widely investigated because of their potential to produce ultra fine-grained microstructures in metals and alloys. Ultra-fine materials exhibit superior mechanical properties such as high strength and ductility. The die for ECAP consists of two channels of equal cross sectional area intersecting at a specified angle that usually varies between 90 and 120 degree. The work-piece undergoes intense plastic deformation within a region in the intersection corner of the two channels. A sound knowledge of the plastic deformation and the load is necessary for understanding the relationships between plastic deformation, grain size and mechanical properties. The friction between the die and the work-piece has great influence on the extrusion pressure and flow in the process. This influence can be studied by Finite Element Modeling. In the present work Finite Element Modeling of ECAP process is carried out using the ABAQUS/Standard software. Channel intersection angle of 105 degree was considered for the study. A cylindrical work-piece of diameter 20 mm and length 105 mm was considered for analysis. Coefficient of friction between the die and billet was varied form 0 to 0.1 in steps of 0.02. The pressure required for deformation v/s the stroke length were obtained and compared with experimental data. The results presented are very useful for understanding the effect of coefficient of friction on extrusion pressure, material flow and strain distribution in the work-piece.

In the present work, dynamic analysis of the model of a typical Indian Railway track system has been done, with special emphasis on "track modulus". The analysis of the track system has been done using the commercially available finite element package ANSYS. A comprehensive static analysis of the track system has been done for validation of the finite element track model. Computation of dynamic track stiffness and track modulus from harmonic loading of the track at different frequencies has also been attempted to predict the track displacement and acceleration responses. These studies have been extended to determine the influence of wheel flat on the track modulus. Such work has not been reported in literature. © Civil-Comp Press, 2008.