• The Journal of the Acoustical Society of Korea
• /
• v.20 no.2
• /
• pp.10-15
• /
• 2001

In this paper, we describe the comparison between the combination of features using a speech and music discrimination, which is classifying between speech and music on audio signals. Audio signals are classified into 3classes (speech, music, speech and music) and 2classes (speech, music). Experiments carried out on three types of feature, Mel-cepstrum, energy, zero-crossings, and try to find a best combination between features to speech and music discrimination. We using a Gaussian Mixture Model (GMM) for discrimination algorithm and combine different features into a single vector prior to modeling the data with a GMM. In 3classes, the best result is achieved using Mel-cepstrum, energy and zero-crossings in a single feature vector (speech: 95.1％, music: 61.9％, speech & music: 55.5％). In 2classes, the best result is achieved using Mel-cepstrum, energy and Mel-cepstrum, energy, zero-crossings in a single feature vector (speech: 98.9％, music: 100％).

• Korean Journal of Mathematics
• /
• v.27 no.1
• /
• pp.9-15
• /
• 2019

In this paper, we study a type of Riemannian manifold, namely quasi Ricci symmetric manifold. Among others, we show that the scalar curvature of a quasi Ricci symmetric manifold is constant. In addition if the manifold is Einstein, then its Ricci tensor is zero. Also we prove that if the associated vector field of a quasi Ricci symmetric manifold is either recurrent or concurrent, then its Ricci tensor is zero.

• Journal of Korea Multimedia Society
• /
• v.17 no.7
• /
• pp.769-777
• /
• 2014

With the rapid interest in Geographic Information System (GIS) contents, a large volume of valuable GIS dataset has been distributed illegally by pirates, hackers, or unauthorized users. Therefore the problem focus on how to protect the copyright of GIS vector map data for storage and transmission. At this point, GIS security techniques focusing on secure network and data encryption have been studied and developed to solve the copyright protection and illegal copy prevention for GIS digital map. But GIS vector map data is very large and current data encryption techniques often encrypt all components of data. That means we have encrypted large amount of data lead to the long encrypting time and high complexity computation. This paper presents a novel selective encryption scheme for GIS vector map data protection to store, transmit or distribute to authorized users using K-means algorithm. The proposed algorithm only encrypts a small part of data based on properties of polylines and polygons in GIS vector map but it can change whole data of GIS vector map. Experimental results verified the proposed algorithm effectively and error in decryption is approximately zero.

• Journal of the Korean Operations Research and Management Science Society
• /
• v.6 no.2
• /
• pp.21-23
• /
• 1981

It is known that two-person zero-sum game with vector payoff can be reduced to a multiple objective linear programming. However, in this case, solutions for the game nay not be one, but many, In many cases in reality, one may need only one solution rather than all solutions. This paper develops a method to find a practical solution for the game by linear programming.

• Architectural research
• /
• v.14 no.4
• /
• pp.143-152
• /
• 2012

A form-finding procedure is presented for planar tensegrity structures. Notably, a simple decision criteria is proposed to select the desirable candidate position vector from the unitary matrix produced by the eigenvalue decomposition of force density matrix. The soundness of the candidate position vector guarantees faster convergence and produces a desirable form of tensegrity without any member having zero-length. Several numerical examples are provided to demonstrate the capability of the proposed form-finding process.

• Journal of the Korean Operations Research and Management Science Society
• /
• v.10 no.2
• /
• pp.28-37
• /
• 1985

An obvious of ESACF approach for model identification by Tsay and Tiao(1984) is that the user may be confused by the elements which are in triangular but marginally larger two standard deviation values. To avoid this drawback, the joint limiting distribution of the vector whose elements are in triangular of ESACF arrays is verified and the statistics to test the nulity of the vector suggested. We illustrate this approach with three examples.

• Proceedings of the KIPE Conference
• /
• 2020.08a
• /
• pp.106-108
• /
• 2020

This paper introduces a new pulse-width modulation (PWM) method to reduce common-mode voltages (CMVs) and compare its performance with other reduced CMV-PWM (RCMV-PWM) methods. To avoid the use of zero-vectors which cause high CMV peaks, the introduced method splits every reference vector into two vectors such that the peak-to-peak magnitude of CMV is reduced by one-third of conventional space-vector PWM (SVPWM). The performance of RCMV-PWMs altered by the modulation index are analyzed with simulation results.

• Bulletin of the Korean Mathematical Society
• /
• v.34 no.2
• /
• pp.273-285
• /
• 1997

The aim of this paper is to present theorems on the exitence of zeros for mappings defined on convex subsets of topological vector spaces with values in a vector space. In addition to natural assumptions of continuity, convexity, and compactness, the mappings are subject to some geometric conditions. In the first theorem, the mapping satisfies a "Darboux-type" property expressed in terms of an auxiliary numerical function. Typically, this functions is, in this case, related to an order structure on the target space. We derive an existence theorem for "obtuse" quasiconvex mappings with values in an ordered vector space. In the second theorem, we prove the existence of a "common zero" for an arbitrary (not necessarily countable) family of mappings satisfying a general "inwardness" condition againg expressed in terms of numerical functions (these numerical functions could be duality pairings (more generally, bilinear forms)). Our inwardness condition encompasses classical inwardness conditions of Leray-Schauder, Altman, or Bergman-Halpern types.

• Proceedings of the KIEE Conference
• /
• 1989.11a
• /
• pp.264-267
• /
• 1989

In the vector controlled induction machine drives, mechanical speed sensors such as shaft encoder and resolver have been used. However, the mechanical speed sensors present some problems and restrict the wide applications of high performance AC drives. This paper describes the vector control strategy with the speed estimation algorithm in which motor slip frequency is calculated. Also, the angle deviation of the rotor flux vector is calculated and instaneously compensated to keep the q axis flux zero in the rotationary reference frame.

• Journal of Institute of Control, Robotics and Systems
• /
• v.9 no.9
• /
• pp.692-699
• /
• 2003

A novel rotor speed estimation method using Model Reference Adaptive Control(MRAC) is proposed to improve the performance of a sensorless vector controller. In the proposed mettled, the stator current is used as the model variable for estimating the speed. In conventional MRAC methods, the relation between the two model errors and the speed estmation error is unclear. Yet, in the proposed method, the stator current error is represented as a function of the first degree for the error value in the speed estimation. Therefore, the proposed method can produce a fast speed estimation and is robust to the parameters error In addition, the proposed method of offers a considerable improvement in the performance of a sensorless vector controller at a low speed. The superiority of the proposed method is verified by simulation and experiment in a low speed region and at a zero-speed.