Showing papers by "Tata Institute of Fundamental Research published in 1987"

A speech enhancement method based on Kalman filtering

Abstract: In this paper, the problem of speech enhancement when only corrupted speech signal is available for processing is considered. For this, the Kalman filtering method is studied and compared with the Wiener filtering method. Its performance is found to be significantly better than the Wiener filtering method. A delayed-Kalman filtering method is also proposed which improves the speech enhancement performance of Kalman filter further.

Equations defining schubert varieties and frobenius splitting of diagonals

Abstract: Let G,(V) be the Grassmannian of r-dimensional linear subspaces of a vector space V. We can identify G,(V) with the decomposable vectors in P A' V, the projective space of lines in the r-th exterior power of V. Thus G,(V) is a projective variety with a natural embedding in P A ' V. A basis el, . . . , e, of V gives the basis eq ^ . . . ^ e~,, i x < i S < . . . < i,, of A' V. The coordinates of a vector in A" V with respect to this basis are called its Plticker coordinates. For a vector in A" V to be decomposable its PRicker coordinates should satisfy certain quadratic relations. For example when n = 4 and r = 2 (the case of projective lines in projective 3-space) the Grassmannian is actually a quadric surface in ps. Hodge gave certain natural quadratic relations in the Pliicker coordinates which define the Grassmarmian in P A ' V (see [5]). In other words, if I is the homogeneous ideal of G,(V) in the homogeneous coordinate ring of P A' V then I is generated by certain quadrics. In developing the theory of spinors, Cartan observed that there is a natural bijection between " pure " spinors and the maximal isotropic spaces of the basic quadratic form. He obtained quadratic relations defining pure spinors in the space of all spinors (see [20]). These results are analogous to the results of Hodge when one deals with the orthogonal group instead of the general linear group. The Grassmannian has a natural decomposition into affine spaces. Let 0 = V 0 C V x C . . . C V . = V be a full flag of linear subspaces with d i m V ~ = i . Let 1 ~< i 1 < i2 < 9 9 9 < i, ~< n be a sequence of integers. Then S = S(ix, . . . , i,) = { W e G,(V) [ dim W ra V ~ / = j }

Regional variations in oral submucous fibrosis in India.

Abstract: Regional variations in the characteristics of submucous fibrosis were studied in two districts in India. In Pune district this condition involved soft palate, uvula and retromolar areas significantly more often than in Ernakulam district. The tongue, floor of the mouth and the hard palate were not involved in Pune. The age of the patients in Pune district was lower than in Ernakulam district. Associated oral cancer, leukoplakia and petechiae were observed solely among patients in Ernakulam. The most important etiologic factor for submucous fibrosis is the chewing of areca nut, and in both areas studied all patients chewed areca nut. In Pune, cured areca nut without other ingredients was chewed by 66% and in Ernakulam, raw areca nut was chewed as an ingredient of pan with tobacco by 100%. Thus in Ernakulam, the juice and the quid were mostly spat out, whereas in Pune they were swallowed. The regional variations in the characteristics of submucous fibrosis could be related to the differences in the areca nut chewing habit between the two areas.

Multi-cones over Schubert varieties

Abstract: Let ~q~ 1 . . . . , 2 ' , be invertible sheaves on a variety X. We may form the multigraded ring A = ~F(X, | where the vector (m) ranges in N". Then C = Spec(A) is a multi-cone. If X is a projective Cohen-Macaulay variety, n = l and L~'l is very ample, Serre has given a criterion for C to be Cohen-Macaulay in terms of the cohomology of sheaves on X. We know no reasonable analog of this result when n > 1. On the other hand we will give such a criterion for C to have rational singularity when X does. This is a stronger restriction on the singularities of C. Then we will give an application to multi-cones based on Schubert varieties. Let G be a reductive algebraic group, B a Borel subgroup and X ~G/B a Schubert variety. Let s176 &~ ..., 5q~ be the line bundles on G/B corresponding to the fundamental weights. An arbitrary line bundle corresponding to a dominant weight has the form

Ionization of xenon by electrons: Partial cross sections for single, double, and triple ionization.

Abstract: High-sensitivity measurements of relative partial cross sections for single, double, and triple ionization of Xe by electron impact have been carried out in the energy region from threshold to 100 eV using a crossed-beam apparatus incorporating a quadrupole mass spectrometer. The weighted sum of the relative partial cross sections at 50 eV are normalized to the total ionization cross section of Rapp and Englander-Golden to yield absolute cross-section functions. Shapes of the partial cross sections for single and double ionization are difficult to account for within a single-particle picture. Comparison of the ${\mathrm{Xe}}^{+}$ data with 4d partial photoionization cross-section measurements indicates the important role played by many-body effects in describing electron-impact ionization of high-Z atoms.

A gene affecting the specificity of the chemosensory neurons of Drosophila.

Abstract: The sensilla on the proboscis and tarsi of Drosophila contain five neurons, four chemosensory and one mechanosensory. The sugar-sensitive neuron, designated S, carries independent acceptor sites for pyranose, furanose and trehalose. Two others, L1 and L2, respond to salts. The fourth neuron, W, is inhibited by salts and sugars, and is believed to mediate detection of water. We describe here a gene in which mutations alter the neurons in such a way that the S cell is excited by salts. As a result, the mutant flies are strongly attracted by NaCl at concentrations which are repellent to the wild type. To our knowledge, this is the first instance of a mutation which changes the specificity of the chemosensory neurons.

The expansion and cosmological evolution of powerful radio sources

Abstract: A simple analytical model is formulated to study beam dynamics in powerful extragalactic radio sources. Two phases are considered for the beam propagation: first through the gaseous halo of the parent galaxy, and then through a much hotter intergalactic medium (IGM). These two media are conceived to be pressure-matched at their interface, which is expected to move closer to the parent galaxy with increasing redshift, owing to a steeply rising IGM pressure. It is argued that the advance of the beam can be ignored quantitatively (notwithstanding a continuing nuclear activity) once its rate has become subsonic relative to the ambient medium. For the input parameters to the model typical values for a powerful radio source, its associated X-ray emitting halo, and the IGM are adopted, and it is shown that the model can explain in a fairly natural way the observed linear sizes of the sources at small redshifts and also their cosmological evolution. The possible relevance of the model to the reported prevalence of subgalactic-sized sources at very high redshifts (z greater than about 2) is also discussed.

IntersectionR-torsion and analytic torsion for pseudomanifolds

Abstract: Pour une pseudovariete X n on definit un invariant Iτ P , la R-torsion intersection, parametrisee par la perversite p. Cet invariant est analogue a la torsion de Reidemeister et est un invariant combinatoire

OH and CH continuous opacity in solar and stellar atmospheres

Abstract: Continuous absorption cross sections of OH and CH have been computed for the temperature range 1000 K to 9000 K. Both OH and CH produce significant ultraviolet opacity in the sun and cool stars. CH is also significant in the visible at 400 nm.

The Angular Size — Redshift Relation as a Cosmological Tool

Abstract: The angular size — redshift (θ − z) relation can in principle be used to discriminate between world models because the angular size subtended by a rigid rod is quite a sensitive function of cosmology, specially at z ≳ 0.5. The test is simpler to apply to objects for which a metric diameter is measured than to objects with isophotal diameters (Sandage 1961). It was first suggested by Hoyle (1938) at the Paris symposium on Radio Astronomy, that the separation between the two lobes of extragalactic radio sources such as Cyg-A, could be used for performing such a test. In an Einstein-de Sitter Universe sources like Cyg-A cannot have angular sizes ≲ 15 arcsec (the minimum occuring at z = 1.25) whereas in the Steady State Universe their sizes should asymptotically approach a value near 4 arcsec at large redshifts. It was not until the early seventies that the test was actually applied to samples of radio quasars with redshifts of upto ~ 2 (Legg 1970; Wiley 1971; Wardle & Wiley 1974). The angular sizes were found to show a large scatter due to a wide distribution of physical sizes and the projection effects associated with the essentially linear radio structures. The upper envelope to the θ -values (which would be expected to show much less scatter) nevertheless appeared to fall off monotonically with increasing z, more or less like the Euclidean relation θ ∝ z-1. The θ − z test thus appeared to be incompatible with the predictions of uniform world models in which the linear sizes of quasars are independent of epoch.

The European hybrid spectrometer

Abstract: The final version of the European hybrid spectrometer (EHS) is described. EHS was used in the NA27 experiment to study charm particles produced in 360 GeV c π-p and 400 GeV c pp interactions. A rapid cycling, high resolution bubble chamber (LEBC) produced bubbles of 17 μm diameter which were measured with a precision of 2.5 μm. Short-lived charm particles are observed with an efficiency up to 90% (D±). The spectrometer accepts 70% of all particles produced, including approximately 100% of those in the forward hemisphere and the momenta of charged particles are measured to better than 1% over the whole momentum range. The charged particle identification detectors give information on 84% of the spectrometer tracks, half of which obtain unique identification. Two lead glass detectors measure the energy of electromagnetic showers such that the π0 mass distribution has a fwhm of 20 MeV c2. Hadronic showers are measured with two iron/scintillator calorimeters. The mass resolutions are 3.0, 0.9 and 11 MeV c2 for K0, Λ0 and D- charged particles respectively. The D0 mass is measured to be (1865.1 ± 1.8 ± 0.8) MeV c2. © 1987.

A Modal Logic for a Subclass of Event Structures

Abstract: This paper introduces a non-interleaved model for the behaviour of distributed computing systems, and an accompanying temporal logic with an explicit treatment of concurrency (based on a notion of local rather than global states). A subclass of event structures (called n-agent event structures) is used as the underlying model -- intended to describe the computational behaviour of n communicating, sequential (and possibly non-deterministic) agents. The logic is centered around indexed modalities to describe the states of knowledge of the individual agents during such a computation. An axiom system for the logic is presented, and a full proof of its soundness and completeness (Henkin style proof) is given.

Signed graphs represented by D

Abstract: In this paper, a characterization of the family of sigraphs represented by D n for all n is proved. The minimal forbidden sigraphs for that family are described. Using that it is shown that any minimal forbidden sigraph for the family of sigraphs with eigenvalues ≥ - 2 has at most 10 vertices. An application to quadratic forms is also given.

Characterization of the valence transition in Yb x In 1-x Cu 2 and anomalous 115 In Knight-shift behavior

Abstract: $^{115}\mathrm{In}$ nuclear-magnetic-resonance Knight shift studies in ${\mathrm{Yb}}_{\mathrm{x}}$${\mathrm{In}}_{1\mathrm{\ensuremath{-}}\mathrm{x}}$${\mathrm{Cu}}_{2}$ microscopically confirm the existence of an anomalous valence transition of Yb ions for some values of x. The rate of variation, as well as the overall variation of the Knight shifts is the largest ever observed in any Yb system. The Knight shifts do not track the susceptibility at low temperatures, the possible implications of which are discussed. Specific-heat results prove the intermediate-valence character of Yb ions in the ground state of these compounds.