• 한국음향학회:학술대회논문집
• /
• 한국음향학회 1999년도 학술발표대회 논문집 제18권 2호
• /
• pp.115-118
• /
• 1999

One of the typical methods to design a codebook is K-means algorithm. This algorithm has the drawbacks that converges to a locally optimal codebook and its performance is mainly decided by an initial codebook. D. Lee's method is almost same as the K-means algorithm except for a modification of a distance value. Those methods have a fixed distance value during all iterations. After many iterations. because the distance between new codevectors and old codevectors is much shorter than the distance in the early stage of iterations, the new codevectors are not affected by distance value. But new codevectors decided in the early stage of learning iterations are much affected by distance value. Therefore it is not appropriate to fix the distance value during all iterations. In this paper, we propose a new algorithm using each different distance value between codevectors for a limited iterations in the early stage of learning iteration. In the experiment, the result show that the proposed method can design better codebooks than the conventional K-means algorithms.

• 전자공학회논문지A
• /
• 제28A권12호
• /
• pp.73-87
• /
• 1991

This paper presents a new approach to the scheduling problem in the high level synthesis. In this approach, iterative rescheduling processes starting with ASAP(As Soon As Possible) scheduling result are performed in a branch-and-bound manner so to arrive at the scheduling result of the lowest hardware cost under the given timing constraint. At each iteration step, only the selected nodes are considered for rescheduling, and the lower bound cost estimation is performed to avoid the unnecessary attempts to search for an optimal result. This branch-and-bound method turns out to be effective in pruning the search space, and thus reducing run time considerably in many cases.

• 대한전자공학회:학술대회논문집
• /
• 대한전자공학회 2001년도 제14회 신호처리 합동 학술대회 논문집
• /
• pp.669-672
• /
• 2001

We propose an efficient algorithm for an interactive content-based image retrieval using relevance feedback. In the proposed algorithm, a new query feature vector first is yielded from the average feature vector of the relevant images that is fed back from the result images of the previous retrieval. Each component weight of a feature vector is computed from an inverse of standard deviation for each component of the relevant images. The updated feature vector of the query and the component weights are used in the iterative retrieval process. In addition, the irrelevant images are excluded from object images in the next iteration to obtain additional performance improvement. In order to evaluate the retrieval performance of the proposed method, we experiment for three image databases, that is, Corel, Vistex, and Ultra databases. We have chosen wavelet moments, BDIP and BVLC, and MFS as features representing the visual content of an image. The experimental results show that the proposed method yields large precision improvement.

• 한국전산유체공학회:학술대회논문집
• /
• 한국전산유체공학회 2000년도 추계 학술대회논문집
• /
• pp.129-134
• /
• 2000

An efficient numerical method to solve the unsteady incompressible Navier-Stokes equations is developed. A fully implicit time advancement is employed to avoid the CFL(Courant-Friedrichs-Lewy) restriction, where the Crank-Nicholson discretization is used for both the diffusion and convection terms. Based on a block LU decomposition, velocity-pressure decoupling is achieved in conjunction with the approximate factorization. Main emphasis is placed on the additional decoupling of the intermediate velocity components with only n th time step velocity The temporal second-order accuracy is Preserved with the approximate factorization without any modification of boundary conditions. Since the decoupled momentum equations are solved without iteration, the computational time is reduced significantly. The present decoupling method is validated by solving the turbulent minimal channel flow unit.

• 전자공학회논문지B
• /
• 제33B권3호
• /
• pp.1-8
• /
• 1996

This paper proposes a novel transversal filter and an optimal LMS algorithm, and show how these can be realized as an adaptive equalizer. The transversal filter consists of a walsh and block pulse functions. in the LMS algorithm with equalizers, the convergence factor is an improtant design parameter because it governs stability and convergence speed. The conventional adaptation techniques use a fixed time constant convergence factor by the trial and error method. In this paper, an optimal method in the choice of the convergence factor is proposed. The proposed algorithm is obtrained that is tailored for each filter tap and is updated at each iteration. The performance of the proposed algorithm is compared iwth those of the conventional TDL and DFT equalizers by computer simulations.

• 한국소음진동공학회:학술대회논문집
• /
• 한국소음진동공학회 2007년도 추계학술대회논문집
• /
• pp.348-351
• /
• 2007

This paper investigates the natural modes of moduled floating structure with module unit connector. As an example structure, a floating parking place($120m{\times}60m$) is considered. In the evaluation of natural modes, numerical equations are formulated by FEM(Finite Element Method) and the natural modes are solved by the subspace iteration method. By comparing results for various stiffness of module unit connector, the effect of stiffness of unit connector is examined.

• 한국소음진동공학회논문집
• /
• 제18권3호
• /
• pp.356-360
• /
• 2008

This paper investigates the natural modes of moduled floating structure with module unit connector. As an example structure, a floating parking place($120m\;{\times}\;60m$) is considered. In the evaluation of natural modes, numerical equations are formulated by FEM(finite element method) and the natural modes are solved by the subspace iteration method. By comparing results for various sizes of module unit, the effect of unit size is investigated. By comparing results for various stiffness of module unit connector, the effect of stiffness of unit connector is also examined.

• 한국정밀공학회지
• /
• 제17권4호
• /
• pp.163-172
• /
• 2000

The postbuckling behavior and progressive damage of composite laminated cylindrical shell under uniform external pressure were investigated by nonlinear finite element method programming. For the finite element analysis, nine-node 3-D degenerated elements were utilized, and arc-length method including line search was adopted for the iteration and load-increment along postbuckling equilibrium path. As results. buckling load, postbucking behavior, and progressive failure f3r various composite laminated cylindrical shells were discussed.

• 한국소성가공학회:학술대회논문집
• /
• 한국소성가공학회 1992년도 춘계학술대회 논문집 92
• /
• pp.159-176
• /
• 1992

Limit analysis has been rendered versatile in many problems such as structural problems and metal forming problems. In metal forming analysis, a slip-line method and an upper bound method approach to limit solutions is considered as the most challenging areas. In the present work, a general algorithm for limit solutions of plastic flow is developed with the use of finite element limit analysis. The algorithm deals with a generalized Holder inequality, a duality theorem, and a combined smoothing and successive approximation in addition to a general procedure for finite element analysis. The algorithm is robust such that from any initial trial solution, the first iteration falls into a convex set which contains the exact solution(s) of the problem. The idea of the algorithm for limit solution is extended from rigid/perfectly-plastic materials to work-hardening materials by the nature of the limit formulation, which is also robust with numerically stable convergence and highly efficient computing time.

• 소성∙가공
• /
• 제16권2호
• /
• pp.138-143
• /
• 2007

Superplastic forming/diffusion bonding (SPF/DB) processes were analyzed using a 3-D rigid visco-plastic finite element method. A constant-triangular element based on membrane approximation and an incremental theory of plasticity are employed for the formulation. The coulomb friction law is used for interface friction between tool and material. Pressure-time relationship for a given optimal strain rate is calculated by stress and pressure values at the previous iteration step. In order to improve the contact searching, hierarchical search algorithm has been applied and implemented into the code. Various geometries including sandwich panel and 3 sheet shape for 3-D SPF/DB model are analyzed using the developed program. The validity fer the analysis is verified by comparison between analysis and results in the literature.