Comparison of HMM and SDTW for Tamil handwritten character recognition

In this paper, we compare the experimental results for Tamil online handwritten character recognition using HMM and Statistical Dynamic Time Warping (SDTW) as classifiers. HMM was used for a 156-class problem. Different feature sets and values for the HMM states & mixtures were tried and the best combination was found to be 16 states & 14 mixtures, giving an accuracy of 85%. The features used in this combination were retained and a SDTW model with 20 states and single Gaussian was used as classifier. Also, the symbol set was increased to include numerals, punctuation marks and special symbols like $, & and #, taking the number of classes to 188. It was found that, with a small addition to the feature set, this simple SDTW classifier performed on par with the more complicated HMM model, giving an accuracy of 84%. Mixture density estimation computations was reduced by 11 times. The recognition is writer independent, as the dataset used is quite large, with a variety of handwriting styles.

On Generalized Reed-Muller Expansions

The generalized Reed-Muller expansions of a switching function are generated using a single Boolean matrix and step-by-step shifting of the principal column.

Determination of the follow current in surge diverters

Using a suitable mathematical model, computations of power/follow current in surge diverters (lightning arresters) have been made from the known short-circuit capacity of the power-frequency source and the nonlinear resistor characteristics. Also the effect of the initiation angle is studied. Typical verifications with the available data have been carried out. The influence of arc drop in the surge-diverter spark gap is neglected.

Minimal unsatisfiable sets: Classification and bounds

Proving the unsatisfiability of propositional Boolean formulas has applications in a wide range of fields. Minimal Unsatisfiable Sets (MUS) are signatures of the property of unsatisfiability in formulas and our understanding of these signatures can be very helpful in answering various algorithmic and structural questions relating to unsatisfiability. In this paper, we explore some combinatorial properties of MUS and use them to devise a classification scheme for MUS. We also derive bounds on the sizes of MUS in Horn, 2-SAT and 3-SAT formulas.

Enhancing Speedup in Network Processing Applications by Exploiting Instruction Reuse with Flow Aggregation

Instruction reuse is a microarchitectural technique that improves the execution time of a program by removing redundant computations at run-time. Although this is the job of an optimizing compiler, they do not succeed many a time due to limited knowledge of run-time data. In this paper we examine instruction reuse of integer ALU and load instructions in network processing applications. Specifically, this paper attempts to answer the following questions: (1) How much of instruction reuse is inherent in network processing applications?, (2) Can reuse be improved by reducing interference in the reuse buffer?, (3) What characteristics of network applications can be exploited to improve reuse?, and (4) What is the effect of reuse on resource contention and memory accesses? We propose an aggregation scheme that combines the high-level concept of network traffic i.e. "flows" with a low level microarchitectural feature of programs i.e. repetition of instructions and data along with an architecture that exploits temporal locality in incoming packet data to improve reuse. We find that for the benchmarks considered, 1% to 50% of instructions are reused while the speedup achieved varies between 1% and 24%. As a side effect, instruction reuse reduces memory traffic and can therefore be considered as a scheme for low power.

Cooperative Profit Sharing in Coalition-Based Resource Allocation in Wireless Networks

We consider a network in which several service providers offer wireless access to their respective subscribed customers through potentially multihop routes. If providers cooperate by jointly deploying and pooling their resources, such as spectrum and infrastructure (e.g., base stations) and agree to serve each others' customers, their aggregate payoffs, and individual shares, may substantially increase through opportunistic utilization of resources. The potential of such cooperation can, however, be realized only if each provider intelligently determines with whom it would cooperate, when it would cooperate, and how it would deploy and share its resources during such cooperation. Also, developing a rational basis for sharing the aggregate payoffs is imperative for the stability of the coalitions. We model such cooperation using the theory of transferable payoff coalitional games. We show that the optimum cooperation strategy, which involves the acquisition, deployment, and allocation of the channels and base stations (to customers), can be computed as the solution of a concave or an integer optimization. We next show that the grand coalition is stable in many different settings, i.e., if all providers cooperate, there is always an operating point that maximizes the providers' aggregate payoff, while offering each a share that removes any incentive to split from the coalition. The optimal cooperation strategy and the stabilizing payoff shares can be obtained in polynomial time by respectively solving the primals and the duals of the above optimizations, using distributed computations and limited exchange of confidential information among the providers. Numerical evaluations reveal that cooperation substantially enhances individual providers' payoffs under the optimal cooperation strategy and several different payoff sharing rules.

A novel variable resolution global spectral method on the sphere

A variable resolution global spectral method is created on the sphere using High resolution Tropical Belt Transformation (HTBT). HTBT belongs to a class of map called reparametrisation maps. HTBT parametrisation of the sphere generates a clustering of points in the entire tropical belt; the density of the grid point distribution decreases smoothly in the domain outside the tropics. This variable resolution method creates finer resolution in the tropics and coarser resolution at the poles. The use of FFT procedure and Gaussian quadrature for the spectral computations retains the numerical efficiency available with the standard global spectral method. Accuracy of the method for meteorological computations are demonstrated by solving Helmholtz equation and non-divergent barotropic vorticity equation on the sphere. (C) 2011 Elsevier Inc. All rights reserved.

Unitary invariants for Hilbert modules of finite rank

We associate a sheaf model to a class of Hilbert modules satisfying a natural finiteness condition. It is obtained as the dual to a linear system of Hermitian vector spaces (in the sense of Grothendieck). A refined notion of curvature is derived from this construction leading to a new unitary invariant for the Hilbert module. A division problem with bounds, originating in Douady's privilege, is related to this framework. A series of concrete computations illustrate the abstract concepts of the paper.

Arbitrary Lagrangian-Eulerian finite-element method for computation of two-phase flows with soluble surfactants

A finite-element scheme based on a coupled arbitrary Lagrangian-Eulerian and Lagrangian approach is developed for the computation of interface flows with soluble surfactants. The numerical scheme is designed to solve the time-dependent Navier-Stokes equations and an evolution equation for the surfactant concentration in the bulk phase, and simultaneously, an evolution equation for the surfactant concentration on the interface. Second-order isoparametric finite elements on moving meshes and second-order isoparametric surface finite elements are used to solve these equations. The interface-resolved moving meshes allow the accurate incorporation of surface forces, Marangoni forces and jumps in the material parameters. The lower-dimensional finite-element meshes for solving the surface evolution equation are part of the interface-resolved moving meshes. The numerical scheme is validated for problems with known analytical solutions. A number of computations to study the influence of the surfactants in 3D-axisymmetric rising bubbles have been performed. The proposed scheme shows excellent conservation of fluid mass and of the total mass of the surfactant. (C) 2012 Elsevier Inc. All rights reserved.

Percolation and connectivity in AB random geometric graphs

Given two independent Poisson point processes Phi((1)), Phi((2)) in R-d, the AB Poisson Boolean model is the graph with the points of Phi((1)) as vertices and with edges between any pair of points for which the intersection of balls of radius 2r centered at these points contains at least one point of Phi((2)). This is a generalization of the AB percolation model on discrete lattices. We show the existence of percolation for all d >= 2 and derive bounds fora critical intensity. We also provide a characterization for this critical intensity when d = 2. To study the connectivity problem, we consider independent Poisson point processes of intensities n and tau n in the unit cube. The AB random geometric graph is defined as above but with balls of radius r. We derive a weak law result for the largest nearest-neighbor distance and almost-sure asymptotic bounds for the connectivity threshold.

Reactivity of Bispropargyl Sulfones under Basic Conditions: Interplay Between Garratt-Braverman and Schmittel/Myers-Saito Cyclization Pathway

The preference for GarrattBraverman (GB) over MyersSaito (MS) and Schmittel (SCM) cyclizations has recently been demonstrated in sulfones capable of undergoing all three of the processes. As the GB cyclization is a self-quenching process, there is a need to change the selectivity to the non-self-quenching MS or SCM pathway so as to enhance the DNA-cleaving efficiency that operates through the radical-mediated process. Herein we report a conformational constraint-based strategy developed by using computations (M06-2X/6-31+G*) to switch the selectivity from GB to MS/SCM pathway which also results in greater DNA-cleavage activity. The preference for GB could be brought back by easing the constraint with the help of spacers.

Two-dimensional flow past circular cylinders using finite volume lattice Boltzmann formulation

In this article, an extension to the total variation diminishing finite volume formulation of the lattice Boltzmann equation method on unstructured meshes was presented. The quadratic least squares procedure is used for the estimation of first-order and second-order spatial gradients of the particle distribution functions. The distribution functions were extrapolated quadratically to the virtual upwind node. The time integration was performed using the fourth-order RungeKutta procedure. A grid convergence study was performed in order to demonstrate the order of accuracy of the present scheme. The formulation was validated for the benchmark two-dimensional, laminar, and unsteady flow past a single circular cylinder. These computations were then investigated for the low Mach number simulations. Further validation was performed for flow past two circular cylinders arranged in tandem and side-by-side. Results of these simulations were extensively compared with the previous numerical data. Copyright (C) 2011 John Wiley & Sons, Ltd.

Criticality guided energy aware speculation for speculative multithreaded processors

Unending quest for performance improvement coupled with the advancements in integrated circuit technology have led to the development of new architectural paradigm. Speculative multithreaded architecture (SpMT) philosophy relies on aggressive speculative execution for improved performance. However, aggressive speculative execution comes with a mixed flavor of improving performance, when successful, and adversely affecting the energy consumption (and performance) because of useless computation in the event of mis-speculation. Dynamic instruction criticality information can be usefully applied to control and guide such an aggressive speculative execution. In this paper, we present a model of micro-execution for SpMT architecture that we have developed to determine the dynamic instruction criticality. We have also developed two novel techniques utilizing the criticality information namely delaying the non-critical loads and the criticality based thread-prediction for reducing useless computations and energy consumption. Experimental results showing break-up of critical instructions and effectiveness of proposed techniques in reducing energy consumption are presented in the context of multiscalar processor that implements SpMT architecture. Our experiments show 17.7% and 11.6% reduction in dynamic energy for criticality based thread prediction and criticality based delayed load scheme respectively while the improvement in dynamic energy delay product is 13.9% and 5.5%, respectively. (c) 2012 Published by Elsevier B.V.

Stabilized SPH-based simulations of impact dynamics using acceleration-corrected artificial viscosity

Artificial viscosity in SPH-based computations of impact dynamics is a numerical artifice that helps stabilize spurious oscillations near the shock fronts and requires certain user-defined parameters. Improper choice of these parameters may lead to spurious entropy generation within the discretized system and make it over-dissipative. This is of particular concern in impact mechanics problems wherein the transient structural response may depend sensitively on the transfer of momentum and kinetic energy due to impact. In order to address this difficulty, an acceleration correction algorithm was proposed in Shaw and Reid (''Heuristic acceleration correction algorithm for use in SPH computations in impact mechanics'', Comput. Methods Appl. Mech. Engrg., 198, 3962-3974) and further rationalized in Shaw et al. (An Optimally Corrected Form of Acceleration Correction Algorithm within SPH-based Simulations of Solid Mechanics, submitted to Comput. Methods Appl. Mech. Engrg). It was shown that the acceleration correction algorithm removes spurious high frequency oscillations in the computed response whilst retaining the stabilizing characteristics of the artificial viscosity in the presence of shocks and layers with sharp gradients. In this paper, we aim at gathering further insights into the acceleration correction algorithm by further exploring its application to problems related to impact dynamics. The numerical evidence in this work thus establishes that, together with the acceleration correction algorithm, SPH can be used as an accurate and efficient tool in dynamic, inelastic structural mechanics. (C) 2011 Elsevier Ltd. All rights reserved.

Seismic stability of a long unsupported circular tunnel

The stability of a long unsupported circular tunnel (opening) in a cohesive frictional soil has been assessed with the inclusion of pseudo-static horizontal earthquake body forces. The analysis has been performed under plane strain conditions by using upper bound finite element limit analysis in combination with a linear optimization procedure. The results have been presented in the form of a non-dimensional stability number (gamma H-max/c); where H = tunnel cover, c refers to soil cohesion and gamma(max) is the maximum unit weight of soil mass which the tunnel can support without collapse. The results have been obtained for various values of H/D (D = diameter of the tunnel), internal friction angle (phi) of soil, and the horizontal earthquake acceleration coefficient (alpha(h)). The computations reveal that the values of the stability numbers (i) decrease quite significantly with an increase in alpha(h), and (ii) become continuously higher for greater values of H/D and phi. As expected, the failure zones around the periphery of the tunnel becomes always asymmetrical with an inclusion of horizontal seismic body forces. (c) 2012 Elsevier Ltd. All rights reserved.