• Geophysics and Geophysical Exploration
• /
• 제11권2호
• /
• pp.85-92
• /
• 2008

Magnetotelluric (MT) methods are widely applied as an effective exploration technique to geothermal surveys. Two-dimensional (2-D) analysis is frequently used to investigate a complicated subsurface structure in a geothermal region. A 2-D finite-element method (FEM) is usually applied to the MT analysis, but we must pay attention to the accuracy of so-called auxiliary fields. Rodi (1976) proposed an algorithm of improving the accuracy of auxiliary fields, and named it as the MOM method. Because it introduces zeros into the diagonal elements of coefficient matrix of the FEM total equation, a pivoting procedure applied to the symmetrical band matrix makes the numerical solution far less efficient. The MOM method was devised mainly for the inversion analysis, in which partial derivatives of both electric and magnetic fields with respect to model parameters are required. In the case of forward modeling, however, we do not have to resort to the MOM method; there is no need of modifying the coefficient matrix, and the auxiliary fields can be elicited from the regular FEM solution. The computational efficiency of the MOM method, however, can be greatly improved through a sophisticated rearrangement of the total equation.

• Journal of the Institute of Electronics Engineers of Korea SD
• /
• 제47권2호
• /
• pp.106-114
• /
• 2010

We proposed an unified design methodology and verification platform for giga-scale System on Chip (SoC). According to the growth of VLSI integration, the existing RTL design methodology has a limitation of a production gap because a design complexity increases. A verification methodology need an evolution to overcome a verification gap. The proposed platform includes a high level synthesis, and we develop a power-aware verification platform for low power design and verification automation using it's results. We developed a verification automation and power-aware verification methodology based on control and data flow graph (CDFG) and an abstract level language and RTL. The verification platform includes self-checking and the coverage driven verification methodology. Especially, the number of the random vector decreases minimum 5.75 times with the constrained random vector algorithm which is developed for the power-aware verification. This platform can verify a low power design with a general logic simulator using a power and power cell modeling method. This unified design and verification platform allow automatically to verify, design and synthesis the giga-scale design from the system level to RTL level in the whole design flow.

• Journal of the Institute of Electronics Engineers of Korea SC
• /
• 제49권4호
• /
• pp.45-54
• /
• 2012

Location-based services with GPS positioning technology as a key technology, but recognizing the current location through satellite communication is not possible in an indoor location-aware technology, low-power short-range communication is primarily made of the study. Especially, as Chirp Spread Spectrum(CSS) based location-aware approach for low-power physical layer IEEE802.15.4a is selected as a standard, Ranging distance estimation techniques and data transfer speed enhancements have been more developed. It is known that the distance measured by CSS ranging has quite a lot of noise as well as its bias. However, the noise problem can be adjusted by modeling the non-zero mean noise value by a scaling factor which corresponds to the change of magnitude of a measured distance vector. In this paper, we propose a localization system using the CSS signal to measure distance for a mobile node taken a measurement of the exact coordinates. By applying the extended kalman filter and least mean squares method, the localization system is faster, more stable. Finally, we evaluate the reliability and accuracy of the proposed algorithm's performance by the experiment for the realization of localization system.

• Proceedings of the Acoustical Society of Korea Conference
• /
• 한국음향학회 2002년도 하계학술발표대회 논문집 제21권 1호
• /
• pp.115-118
• /
• 2002

Parameters used in a speaker recognition system are desirable expressing speaker's characteristics filly and have in a speech. That is to say, if inter-speaker than intra-speaker variance a big characteristic, it is useful to distinguish between speakers. Also, to make minimum error between speakers, it is required the improved recognition technology as well as the distinguishing characteristics. When we see the result of recent simulation performance, we obtain more exact performance by using dynamic characteristics and constant characteristics by a speaking habit. Therefore we suggest it to solve this problem as followings. The prosodic information is used by a characteristic vector of speech. Characteristics vector generally using in speaker recognition system is a modeling spectrum information and is working for a high performance in non-noise circumstance. However, it is found a problem that characteristic vector is distorted in noise circumstance and it makes a reduction of recognition rate. In this paper, we change pitch line divided by segment which can estimate a dynamic characteristic and it is used as a recognition characteristic. we confirmed that the dynamic characteristic is very robust in noise circumstance with a simulation. We make a decision of acceptance or rejection by comparing test pattern and recognition rate using the proposed algorithm has more improvement than using spectrum and prosodic information. Especially stational recognition rate can be obtained in noise circumstance through the simulation.

• Journal of the Computational Structural Engineering Institute of Korea
• /
• 제23권1호
• /
• pp.81-94
• /
• 2010

This paper deals with the fracture simulation of UHPFRC girder with the interface type model. Based on the existing numerical simulation of quasi-brittle fracture in normal strength concrete, constitutive modeling for UHPFRC I-girder has been improved by including a tensile hardening at the failure surface. The finite element formulation is based on a triangular unit, constructed from constant strain triangles, with nodes along its sides and neither at the vertex nor the center of the unit. Fracture is simulated through a hardening/softening fracture constitutive law in tension, a softening fracture constitutive law in shear as well as in compression at the boundary nodes, with the material within the triangular unit remaining linear elastic. LCP is used to formulate the path-dependent hardening-softening behavior in non-holonomic rate form and a mathematical programming algorithm is employed to solve the LCP. The piece-wise linear inelastic yielding-failure/failure surface is modeled with two compressive caps, two Mohr-Coulomb failure surfaces, a tensile yielding surface and a tensile failure surface. The comparison between test results and numerical results indicates this method effectively simulates the deformation and failure of specimen.

• Phonetics and Speech Sciences
• /
• 제12권2호
• /
• pp.29-37
• /
• 2020

Recently, the neural network-based deep learning algorithm has dramatically improved performance compared to the classical Gaussian mixture model based hidden Markov model (GMM-HMM) automatic speech recognition (ASR) system. In addition, researches on end-to-end (E2E) speech recognition systems integrating language modeling and decoding processes have been actively conducted to better utilize the advantages of deep learning techniques. In general, E2E ASR systems consist of multiple layers of encoder-decoder structure with attention. Therefore, E2E ASR systems require data with a large amount of speech-text paired data in order to achieve good performance. Obtaining speech-text paired data requires a lot of human labor and time, and is a high barrier to building E2E ASR system. Therefore, there are previous studies that improve the performance of E2E ASR system using relatively small amount of speech-text paired data, but most studies have been conducted by using only speech-only data or text-only data. In this study, we proposed a semi-supervised training method that enables E2E ASR system to perform well in corpus in different domains by using both speech or text only data. The proposed method works effectively by adapting to different domains, showing good performance in the target domain and not degrading much in the source domain.

• Journal of Radiation Protection and Research
• /
• 제42권1호
• /
• pp.9-15
• /
• 2017

Background: The purpose of this study is to assign an appropriate density to virtual phantom for 2D diode array detector with different dose calculation algorithms to guarantee the accuracy of patient-specific QA. Materials and Methods: Ten VMAT plans with 6 MV photon beam and ten VMAT plans with 15 MV photon beam were selected retrospectively. The computed tomography (CT) images of MapCHECK2 with MapPHAN were acquired to design the virtual phantom images. For all plans, dose distributions were calculated for the virtual phantoms with four different materials by AAA and AXB algorithms. The four materials were polystyrene, 455 HU, Jursinic phantom, and PVC. Passing rates for several gamma criteria were calculated by comparing the measured dose distribution with calculated dose distributions of four materials. Results and Discussion: For validation of AXB modeling in clinic, the mean percentages of agreement in the cases of dose difference criteria of 1.0% and 2.0% for 6 MV were $97.2%{\pm}2.3%$, and $99.4%{\pm}1.1%$, respectively while those for 15 MV were $98.5%{\pm}0.85%$ and $99.8%{\pm}0.2%$, respectively. In the case of 2%/2 mm, all mean passing rates were more than 96.0% and 97.2% for 6 MV and 15 MV, respectively, regardless of the virtual phantoms of different materials and dose calculation algorithms. The passing rates in all criteria slightly increased for AXB as well as AAA when using 455 HU rather than polystyrene. Conclusion: The virtual phantom which had a 455 HU values showed high passing rates for all gamma criteria. To guarantee the accuracy of patent-specific VMAT QA, each institution should fine-tune the mass density or HU values of this device.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• 제12권2호
• /
• pp.821-828
• /
• 2011

This paper developed the algorithm for improving the performance the auto pilot in the autonomous vehicle system consisting of the Track keeping control, the Automatic steering, and the Automatic mooring control. The automatic steering is the control device that could save the voyage distance and cost of fuel by reducing the unnecessary burden of driving due to the continuous artificial navigation, and avoiding the route deviation. During the step of the ship autonomic navigation control, since the wind power or the tidal force could make the ship deviate from the fixed course, the automatic steering calculates the difference between actual sailing line and the set course to keep the ship sailing in the vicinity of intended course. first, we could get the transfer function for the modeling of ship according to the Nomoto model. Considering the maneuverability, we propose it as linear model with only 4 degree of freedoms to present the heading angle response to the input of rudder angle. In this paper, the model of ship is derived from the simplified Nomoto model. Since the proposed model considers the maximum angle and rudder rate of the ship auto pilot and also designs the Fuzzy controller based on existing PID controller, the performance of the steering machine is well improved.

• The KIPS Transactions:PartD
• /
• 제11D권6호
• /
• pp.1259-1268
• /
• 2004

To organize training in limited training area for an actuai combat, realistic training simulation plugged in by various battle conditions is essential. In this paper, we propose a virtual target overlay technique which does not use a virtual image, but Projects a virtual target on ground-based CCD image by appointed scenario for a realistic training simulation. In the proposed method, we create a realistic 3D model (for an instructor) by using high resolution Geographic Tag Image File Format(GeoTIFF) satellite image and Digital Terrain Elevation Data (DTED), and extract the road area from a given CCD image (for both an instructor and a trainee). Satellite images and ground-based sensor images have many differences in observation position, resolution, and scale, thus yielding many difficulties in feature-based matching. Hence, we propose a moving synchronization technique that projects the target on the sensor image according to the marked moving path on 3D satellite image by applying Thin-Plate Spline(TPS) interpolation function, which is an image warping function, on the two given sets of corresponding control point pair. To show the experimental result of the proposed method, we employed two Pentium4 1.8MHz personal computer systems equipped with 512MBs of RAM, and the satellite and sensor images of Daejoen area are also been utilized. The experimental result revealed the effective-ness of proposed algorithm.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• 제10권11호
• /
• pp.2075-2082
• /
• 2006

This paper presents modeling and optimization techniques for hish efficiency solar cell process on single-crystalline float zone (FZ) wafers. Among a sequence of multiple steps of fabrication, the followings are the most sensitive steps for the contact formation: 1) Emitter formation by diffusion; 2) Anti-reflection-coating (ARC) with silicon nitride using plasma-enhanced chemical vapor deposition (PECVD); 3) Screen-printing for front and back metalization; and 4) Contact formation by firing. In order to increase the performance of solar cells in terms of efficiency, the contact formation process is modeled and optimized using neural networks and genetic algorithms, respectively. This paper utilizes the design of experiments (DOE) in contact formation to reduce process time and fabrication costs. The experiments were designed by using central composite design which consists of 24 factorial design augmented by 8 axial points with three center points. After contact formation process, the efficiency of the fabricated solar cell is modeled using neural networks. Established efficiency model is then used for the analysis of the process characteristics and process optimization for more efficient solar cell fabrication.