1,362,232 research outputs found

    Sequential decision fusion for controlled detection errors

    No full text
    Information fusion in biometrics has received considerable attention. The architecture proposed here is based on the sequential integration of multi-instance and multi-sample fusion schemes. This method is analytically shown to improve the performance and allow a controlled trade-off between false alarms and false rejects when the classifier decisions are statistically independent. Equations developed for detection error rates are experimentally evaluated by considering the proposed architecture for text dependent speaker verification using HMM based digit dependent speaker models. The tuning of parameters, n classifiers and m attempts/samples, is investigated and the resultant detection error trade-off performance is evaluated on individual digits. Results show that performance improvement can be achieved even for weaker classifiers (FRR-19.6%, FAR-16.7%). The architectures investigated apply to speaker verification from spoken digit strings such as credit card numbers in telephone or VOIP or internet based applications

    Carrier-Based Modulation Scheme for Dual Inverter-Fed Asymmetrical Six-Phase Drives Supplied from a Single DC Source

    No full text
    This paper proposes a carrier-based switching strategy for dual inverter-fed asymmetric six-phase motor drives. Unlike traditional pulse width modulation techniques for multilevel inverters, the proposed method uses a single-carrier-based PWM derived by six-phase, two-level inverter-fed, star-connected asymmetrical six-phase motor (ASPM) and is derived from traditional pulse width modulation techniques for multilevel inverters. It exploits the geometric similarity between the resultant space vector (SV) structure across the ASPM stator and the SV structure of a six-phase, two-level VSI. Proposed scheme eliminates dwell time calculation, sector identification and other multilevel SVPWM related operations, the common mode voltage across the stator windings and the net common mode voltage across the two sets of three-phase winding. Moreover, the scheme eliminates non-torque-producing auxiliary plane harmonic voltages and provides a higher stator voltage (up to double) compared to single end fed ASPM drives for a given DC source voltage. In this way, operations from a single DC source without common mode zero sequence currents circulating through the stator windings are allowed. The proposed scheme is validated by extensive MATLAB/Simulink simulations

    Inapproximability of Rainbow Colouring

    No full text
    A rainbow colouring of a connected graph G is a colouring of the edges of G such that every pair of vertices in G is connected by at least one path in which no two edges are coloured the same. The minimum number of colours required to rainbow colour G is called its rainbow connection number. Chakraborty, Fischer, Matsliah and Yuster have shown that it is NP-hard to compute the rainbow connection number of graphs [J. Comb. Optim., 2011]. Basavaraju, Chandran, Rajendraprasad and Ramaswamy have reported an (r+3)-factor approximation algorithm to rainbow colour any graph of radius r [Graphs and Combinatorics, 2012]. In this article, we use a result of Guruswami, Håstad and Sudan on the NP-hardness of colouring a 2-colourable 4-uniform hypergraph using constantly many colours [SIAM J. Comput., 2002] to show that for every positive integer k, it is NP-hard to distinguish between graphs with rainbow connection number 2k+2 and 4k+2. This, in turn, implies that there cannot exist a polynomial time algorithm to rainbow colour graphs with less than twice the optimum number of colours, unless P=NP. The authors have earlier shown that the rainbow connection number problem remains NP-hard even when restricted to the class of chordal graphs, though in this case a 4-factor approximation algorithm is available [COCOON, 2012]. In this article, we improve upon the 4-factor approximation algorithm to design a linear-time algorithm that can rainbow colour a chordal graph G using at most 3/2 times the minimum number of colours if G is bridgeless and at most 5/2 times the minimum number of colours otherwise. Finally we show that the rainbow connection number of bridgeless chordal graphs cannot be polynomial-time approximated to a factor less than 5/4, unless P=NP

    Signal processing to improve target detection using ground penetrating radar

    No full text
    This paper focuses on some important GPR pre-processing tasks. These tasks are necessary to improve formal target detection and estimation stages. Evaluations performed using a conventional matched filter on real GPR data demonstrate the benefits of pre-processing. This effort is conducted with the ultimate goal of realising a reliable GPR-based coal-thickness sensor.

    Going Beyond Counting First Authors in Author Co-citation Analysis

    No full text
    The present study examines one of the fundamental aspects of author co-citation analysis (ACA) - the way co-citation counts are defined. Co-citation counting provides the data on which all subsequent statistical analyses and mappings are based, and we compare ACA results based on two different types of co-citation counting - the traditional type that only counts the first one among a cited work's authors on the one hand and a non-traditional type that takes into account the first 5 authors of a cited work on the other hand. Results indicate that the picture produced through this non-traditional author co-citation counting contains more coherent author groups and is therefore considerably clearer. However, this picture represents fewer specialties in the research field being studied than that produced through the traditional first-author co-citation counting when the same number of top-ranked authors is selected and analyzed. Reasons for these effects are discussed

    A New Species of Lanelater Arnett (Coleoptera: Elateridae) from the Andaman and Nicobar Islands, India

    No full text
    Chandran, Nithya, Dubey, Anil Kumar (2021): A New Species of Lanelater Arnett (Coleoptera: Elateridae) from the Andaman and Nicobar Islands, India. The Coleopterists Bulletin 75 (1): 240-246, DOI: 10.1649/0010-065X-75.1.240, URL: http://dx.doi.org/10.1649/0010-065x-75.1.24

    Stereotaip gender dalam novel terpilih Ramani Chandran / P. Pama A/P Periasamy

    No full text
    Fokus kajian ini adalah pada stereotaip gender dalam novel Tamil. Stereotaip merupakan satu fenomena yang wujud dalam semua masyarakat. Kita boleh menganggap bahawa golongan yang lebih lemah dalam sesebuah masyarakat selalunya distereotaipkan. Golongan ini boleh terdiri daripada lelaki atau wanita. Novel, yang dikatakan sebagai cermin masyarakat digunakan untuk mengkaji stereotaip lelaki dan wanita. Bagi tujuan ini, 8 buah novel yang telah dihasilkan oleh penulis prolifik Ramani Chandran yang mewakili 4 dekad, iaitu dari tahun 1971 hingga tahun 2010 digunakan sebagai data kajian ini. Data kajian telah dianalisis dan dirumuskan bagi manjawab soalan-soalan kajian. Dapatan kajian menunjukkan bahawa sememangnya wujud perubahan unsur-unsur stereotaip lelaki dan wanita dalam penulisan Ramani Chandran. Dalam novel-novel yang dihasilkan pada awal tahun 70-an didapati novelis sangat terpengaruh dengan stereotaip wanita yang sedia wujud dalam masyarakat India. Tetapi persepsi stereotaip novelis terhadap lelaki dan wanita India didapati berubah mengikut peredaran masa. Keadaan ini menggambarkan bahawa stereotaip yang ditentukan oleh sesebuah masyarakat juga berubah mengikut peredaran masa. Ini jelas dilihat dalam novel-novel yang dihasilkan oleh Ramani Chandran, dalam tempoh masa 40 tahun. Ini membuktikan bahawa pelanggaran stereotaip lelaki dan wanita India sudah pasti berlaku dalam sesebuah masyarakat dan pelanggaran-pelanggaran ini terlihat dalam karya-karya sastera, khususnya novel

    A modified peridynamic method to model the fracture behaviour of nanocomposites

    No full text
    In the present work, a numerical investigation based on a modified peridynamic method of fracture properties of epoxy resin reinforced by nanoparticles, more specifically hyperbranched polyester (HBP), was conducted. Due to the specific features of HBP, certain material nodes in the numerical model were constrained to accurately replicate the effect of HBP in pure epoxy resin, and a Monte Carlo method was used to simulate the random distribution of HBP. The numerical model was validated by fracture tests with single-edge-notched bending (SENB) samples. Moreover, the effect of the HBP weight fraction on the fracture properties was analysed. With increaseing HBP weight fraction, a three-stage increase of mode-I fracture toughness in a rapid-slow-rapid manner was obtained. Overall, the proposed modified peridynamic method provided a macroscale analysis of the fracture behaviour of nanocomposites reinforced by HBP in a simulation framework
    corecore