Common Ground Method of Current Injection in Electrical Impedance Tomography

Surface electrodes are essentially required to be switched for boundary data collection in electrical impedance tomography (Ell). Parallel digital data bits are required to operate the multiplexers used, generally, for electrode switching in ELT. More the electrodes in an EIT system more the digital data bits are needed. For a sixteen electrode system. 16 parallel digital data bits are required to operate the multiplexers in opposite or neighbouring current injection method. In this paper a common ground current injection is proposed for EIT and the resistivity imaging is studied. Common ground method needs only two analog multiplexers each of which need only 4 digital data bits and hence only 8 digital bits are required to switch the 16 surface electrodes. Results show that the USB based data acquisition system sequentially generate digital data required for multiplexers operating in common ground current injection method. The profile of the boundary data collected from practical phantom show that the multiplexers are operating in the required sequence in common ground current injection protocol. The voltage peaks obtained for all the inhomogeneity configurations are found at the accurate positions in the boundary data matrix which proved the sequential operation of multiplexers. Resistivity images reconstructed from the boundary data collected from the practical phantom with different configurations also show that the entire digital data generation module is functioning properly. Reconstructed images and their image parameters proved that the boundary data are successfully acquired by the DAQ system which in turn indicates a sequential and proper operation of multiplexers.

A Model for the Temperature Distribution in Resin Impregnated Paper Bushings

Resin impregnated paper (RIP) is a relatively new insulation system recommended for the use in transformer bushings. In the recent past, RIP has acquired prominence as insulation in bushings, over conventional oil impregnated paper (OIP), in view of its overwhelming advantages the more important among them being low dielectric loss and possibility for positioning the bushing at any desired angle over the transformer. In addition, the fact that such systems do not pose problems of fire hazard is counted as a very important consideration. The disadvantage of RIP compared to OIP, however, is its much higher cost and involved manufacturing process. The temperature rise in RIP bushings under normal operating conditions is seen to be a difficult parameter to control in view of the limited options for effective cooling. It is therefore essential to take serious note of this aspect, to arrest rapid deterioration of bushing. The degradation of dry-type insulation such as RIP is often due to thermal stress. The long time performance thereof, depends strongly, on the maximum operating temperature. With this in view, the Authors have developed a theoretical model and computational method to study the temperature distribution in the body of insulation. The Authors consider that the basis for the model as being the temperature and electric stress aided AC conductivity. The ensuing heat balance (continuity) equations in 2-D cylindrical geometry are treated as a Dirichelet-Neumann boundary value problem.

Fast Likelihood Computation in Speech Recognition using Matrices

Acoustic modeling using mixtures of multivariate Gaussians is the prevalent approach for many speech processing problems. Computing likelihoods against a large set of Gaussians is required as a part of many speech processing systems and it is the computationally dominant phase for Large Vocabulary Continuous Speech Recognition (LVCSR) systems. We express the likelihood computation as a multiplication of matrices representing augmented feature vectors and Gaussian parameters. The computational gain of this approach over traditional methods is by exploiting the structure of these matrices and efficient implementation of their multiplication. In particular, we explore direct low-rank approximation of the Gaussian parameter matrix and indirect derivation of low-rank factors of the Gaussian parameter matrix by optimum approximation of the likelihood matrix. We show that both the methods lead to similar speedups but the latter leads to far lesser impact on the recognition accuracy. Experiments on 1,138 work vocabulary RM1 task and 6,224 word vocabulary TIMIT task using Sphinx 3.7 system show that, for a typical case the matrix multiplication based approach leads to overall speedup of 46 % on RM1 task and 115 % for TIMIT task. Our low-rank approximation methods provide a way for trading off recognition accuracy for a further increase in computational performance extending overall speedups up to 61 % for RM1 and 119 % for TIMIT for an increase of word error rate (WER) from 3.2 to 3.5 % for RM1 and for no increase in WER for TIMIT. We also express pairwise Euclidean distance computation phase in Dynamic Time Warping (DTW) in terms of matrix multiplication leading to saving of approximately of computational operations. In our experiments using efficient implementation of matrix multiplication, this leads to a speedup of 5.6 in computing the pairwise Euclidean distances and overall speedup up to 3.25 for DTW.

A Savitzky-Golay Filtering Perspective of Dynamic Feature Computation

We address the classical problem of delta feature computation, and interpret the operation involved in terms of Savitzky- Golay (SG) filtering. Features such as themel-frequency cepstral coefficients (MFCCs), obtained based on short-time spectra of the speech signal, are commonly used in speech recognition tasks. In order to incorporate the dynamics of speech, auxiliary delta and delta-delta features, which are computed as temporal derivatives of the original features, are used. Typically, the delta features are computed in a smooth fashion using local least-squares (LS) polynomial fitting on each feature vector component trajectory. In the light of the original work of Savitzky and Golay, and a recent article by Schafer in IEEE Signal Processing Magazine, we interpret the dynamic feature vector computation for arbitrary derivative orders as SG filtering with a fixed impulse response. This filtering equivalence brings in significantly lower latency with no loss in accuracy, as validated by results on a TIMIT phoneme recognition task. The SG filters involved in dynamic parameter computation can be viewed as modulation filters, proposed by Hermansky.

Device oriented statistical modeling method for process variability in 45nm analog CMOS technology

With the rapid scaling down of the semiconductor process technology, the process variation aware circuit design has become essential today. Several statistical models have been proposed to deal with the process variation. We propose an accurate BSIM model for handling variability in 45nm CMOS technology. The MOSFET is designed to meet the specification of low standby power technology of International Technology Roadmap for Semiconductors (ITRS).The process parameters variation of annealing temperature, oxide thickness, halo dose and title angle of halo implant are considered for the model development. One parameter variation at a time is considered for developing the model. The model validation is done by performance matching with device simulation results and reported error is less than 10%.© (2012) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.

Fast acoustic likelihood computation using low-rank matrix approximation

Acoustic modeling using mixtures of multivariate Gaussians is the prevalent approach for many speech processing problems. Computing likelihoods against a large set of Gaussians is required as a part of many speech processing systems and it is the computationally dominant phase for LVCSR systems. We express the likelihood computation as a multiplication of matrices representing augmented feature vectors and Gaussian parameters. The computational gain of this approach over traditional methods is by exploiting the structure of these matrices and efficient implementation of their multiplication.In particular, we explore direct low-rank approximation of the Gaussian parameter matrix and indirect derivation of low-rank factors of the Gaussian parameter matrix by optimum approximation of the likelihood matrix. We show that both the methods lead to similar speedups but the latter leads to far lesser impact on the recognition accuracy. Experiments on a 1138 word vocabulary RM1 task using Sphinx 3.7 system show that, for a typical case the matrix multiplication approach leads to overall speedup of 46%. Both the low-rank approximation methods increase the speedup to around 60%, with the former method increasing the word error rate (WER) from 3.2% to 6.6%, while the latter increases the WER from 3.2% to 3.5%.

Speeded up Gaussian mixture model algorithm for background subtraction

Adaptive Gaussian Mixture Models (GMM) have been one of the most popular and successful approaches to perform foreground segmentation on multimodal background scenes. However, the good accuracy of the GMM algorithm comes at a high computational cost. An improved GMM technique was proposed by Zivkovic to reduce computational cost by minimizing the number of modes adaptively. In this paper, we propose a modification to his adaptive GMM algorithm that further reduces execution time by replacing expensive floating point computations with low cost integer operations. To maintain accuracy, we derive a heuristic that computes periodic floating point updates for the GMM weight parameter using the value of an integer counter. Experiments show speedups in the range of 1.33 - 1.44 on standard video datasets where a large fraction of pixels are multimodal.

Further results on geometric properties of a family of relative entropies

This paper extends some geometric properties of a one-parameter family of relative entropies. These arise as redundancies when cumulants of compressed lengths are considered instead of expected compressed lengths. These parametric relative entropies are a generalization of the Kullback-Leibler divergence. They satisfy the Pythagorean property and behave like squared distances. This property, which was known for finite alphabet spaces, is now extended for general measure spaces. Existence of projections onto convex and certain closed sets is also established. Our results may have applications in the Rényi entropy maximization rule of statistical physics.

Full-rate full-diversity achieving MIMO precoding with partial CSIT

In this paper, we consider a slow-fading nt ×nr multiple-input multiple-output (MIMO) channel subjected to block fading. Reliability (in terms of achieved diversity order) and rate (in number of symbols transmitted per channel use) are of interest in such channels. We propose a new precoding scheme which achieves both full diversity (nt ×nrth order diversity) as well as full rate (nt symbols per channel use) using partial channel state information at the transmitter (CSIT). The proposed scheme achieves full diversity and improved coding gain through an optimization over the choice of constellation sets. The optimization maximizes dmin2 for our precoding scheme subject to an energy constraint. The scheme requires feedback of nt - 1 angle parameter values, compared to 2ntnr real coefficients in case of full CSIT. Further, for the case of nt × 1 system, we prove that the capacity achieved by the proposed scheme is same as that achieved with full CSIT. Error rate performance results for nt = 3,4,8 show that the proposed scheme performs better than other precoding schemes in the literature; the better performance is due to the choice of the signal sets and the feedback angles in the proposed scheme.

Analysis of distance relay trip regions for EHV and UHV transmission lines

This paper presents comparative evaluation of the distance relay characteristics for UHV and EHV transmission lines. Distance protection relay characteristics for the EHV and UHV systems are developed using Electromagnetic Transients (EMT) program. The variation of ideal trip boundaries for both the systems are presented. Unlike the conventional distance protection relay which uses a lumped parameter model, this paper uses the distributed parameter model. The effect of larger shunt susceptance on the trip boundaries is highlighted. Performance of distance relay with ideal trip boundaries for EHV and UHV lines have been tested for various fault locations and fault resistances. Electromagnetic Transients (EMT) program has been developed considering distributed parameter line model for simulating the test systems. The voltage and current phasors are computed from the signals using an improved full cycle DFT algorithm taking 20 samples per cycle. Two practical transmission systems of Indian power grid, namely 765 kV UHV transmission line and SREB 24-bus 400kV EHV system are used to test the performance of the proposed approach.

Efficient post-processing techniques for speech enhancement

We address the problem of speech enhancement in real-world noisy scenarios. We propose to solve the problem in two stages, the first comprising a generalized spectral subtraction technique, followed by a sequence of perceptually-motivated post-processing algorithms. The role of the post-processing algorithms is to compensate for the effects of noise as well as to suppress any artifacts created by the first-stage processing. The key post-processing mechanisms are aimed at suppressing musical noise and to enhance the formant structure of voiced speech as well as to denoise the linear-prediction residual. The parameter values in the techniques are fixed optimally by experimentally evaluating the enhancement performance as a function of the parameters. We used the Carnegie-Mellon university Arctic database for our experiments. We considered three real-world noise types: fan noise, car noise, and motorbike noise. The enhancement performance was evaluated by conducting listening experiments on 12 subjects. The listeners reported a clear improvement (MOS improvement of 0.5 on an average) over the noisy signal in the perceived quality (increase in the mean-opinion score (MOS)) for positive signal-to-noise-ratios (SNRs). For negative SNRs, however, the improvement was found to be marginal.

Impact of chemical treatment on the surface, structure, optical and electrical properties of SnS thin films

The impact of chemical treatment on the surface morphology and other physical properties of tin monosulphide (SnS) thin films have been investigated. The SnS films treated with selected organic solvents exhibited strong improvement in their crystalline-quality and considerable decrease in electrical resistivity. Particularly, the films treated with chloroform showed very low electrical resistivity of similar to 5 Omega cm and a low optical band gap of 1.81 eV as compared to untreated and treated SnS films with other chemicals. From these studies we realized that the chemical treatment of SnS films has strong impact on their surface morphology and also on other physical properties. (C) 2012 Elsevier B.V. All rights reserved.

Electrical switching studies on amorphous Ge‐Te‐Sn thin films

Electrical switching studies on amorphous Ge17Te83−xSnx thin films (1 ≤ x ≤ 4) has been done to find their suitability for Phase Change Memory application; Bulk ingots in glassy form are prepared using conventional melt quenching technique and the thin films are coated using flash evaporation technique. Samples are found to exhibit memory type of electrical switching behavior. The switching voltages of Ge17Te83−xSnx thin films have been found to decrease with increase in Sn concentration. The comparatively lower switching voltages of Ge17Te83−xSnx samples, make them suitable candidates for phase change memory applications.

Internal defect and process parameter analysis during friction stir welding of Al 6061 sheets

Welding parameters like welding speed, rotation speed, plunge depth, shoulder diameter etc., influence the weld zone properties, microstructure of friction stir welds, and forming behavior of welded sheets in a synergistic fashion. The main aims of the present work are to (1) analyze the effect of welding speed, rotation speed, plunge depth, and shoulder diameter on the formation of internal defects during friction stir welding (FSW), (2) study the effect on axial force and torque during welding, (c) optimize the welding parameters for producing internal defect-free welds, and (d) propose and validate a simple criterion to identify defect-free weld formation. The base material used for FSW throughout the work is Al 6061T6 having a thickness value of 2.1 mm. Only butt welding of sheets is aimed in the present work. It is observed from the present analysis that higher welding speed, higher rotation speed, and higher plunge depth are preferred for producing a weld without internal defects. All the shoulder diameters used for FSW in the present work produced defect-free welds. The axial force and torque are not constant and a large variation is seen with respect to FSW parameters that produced defective welds. In the case of defect-free weld formation, the axial force and torque are relatively constant. A simple criterion, (a,tau/a,p)(defective) > (a,tau/a,p)(defect free) and (a,F/a,p)(defective) > (a,F/a,p)(defect free), is proposed with this observation for identifying the onset of defect-free weld formation. Here F is axial force, tau is torque, and p is welding speed or tool rotation speed or plunge depth. The same criterion is validated with respect to Al 5xxx base material. Even in this case, the axial force and torque remained constant while producing defect-free welds.

A LSQR-type method provides a computationally efficient automated optimal choice of regularization parameter in diffuse optical tomography

Purpose: Developing a computationally efficient automated method for the optimal choice of regularization parameter in diffuse optical tomography. Methods: The least-squares QR (LSQR)-type method that uses Lanczos bidiagonalization is known to be computationally efficient in performing the reconstruction procedure in diffuse optical tomography. The same is effectively deployed via an optimization procedure that uses the simplex method to find the optimal regularization parameter. The proposed LSQR-type method is compared with the traditional methods such as L-curve, generalized cross-validation (GCV), and recently proposed minimal residual method (MRM)-based choice of regularization parameter using numerical and experimental phantom data. Results: The results indicate that the proposed LSQR-type and MRM-based methods performance in terms of reconstructed image quality is similar and superior compared to L-curve and GCV-based methods. The proposed method computational complexity is at least five times lower compared to MRM-based method, making it an optimal technique. Conclusions: The LSQR-type method was able to overcome the inherent limitation of computationally expensive nature of MRM-based automated way finding the optimal regularization parameter in diffuse optical tomographic imaging, making this method more suitable to be deployed in real-time. (C) 2013 American Association of Physicists in Medicine. http://dx.doi.org/10.1118/1.4792459]