• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.32 no.3C
• /
• pp.338-347
• /
• 2007

In this paper, a weighted block adaptive channel estimation (WBA-CE) for a space-time block-coded (STBC) single-carrier transmission with a cyclic-prefix is proposed. In operation of the WBA-CE, a STBC matrix-wise block for filter input symbols is first formulated. Applying a weighted a posteriori error vector-based least-square (LS) criterion for this block, the coefficient correction terms of the WBA-CE are then computed. An approximate steady-state excess mean-square error (EMSE) of the WBA-CE for the stationary optimal coefficient is also analyzed. Simulation results show in a time-varying typical urban (TU) channel that the proposed channel estimator provides better bit-error-rate (BER) performances than conventional algorithms such as the NLMS and RLS channel estimators.

• Journal of the Korean Mathematical Society
• /
• v.49 no.3
• /
• pp.549-569
• /
• 2012

In this paper asymptotic error expansions for mixed finite element approximations to a class of second order elliptic optimal control problems are derived under rectangular meshes, and the Richardson extrapolation of two different schemes and interpolation defect correction can be applied to increase the accuracy of the approximations. As a by-product, we illustrate that all the approximations of higher accuracy can be used to form a class of a posteriori error estimators of the mixed finite element method for optimal control problems.

• Journal of the Korean Institute of Telematics and Electronics B
• /
• v.31B no.12
• /
• pp.108-115
• /
• 1994

Most pattern classifiers have been designed based on the ML (Maximum Likelihood) training algorithm which is simple and relatively powerful. The ML training is an efficient algorithm to individually estimate the model parameters of each class under the assumption that all class models in a classifier are statistically independent. That assumption, however, is not valid in many real situations, which degrades the performance of the classifier. In this paper, we propose a minimum-error-rate training algorithm based on the MAP (Maximum a Posteriori) approach. The algorithm regards the normalized outputs of the classifier as estimates of the a posteriori probability, and tries to maximize those estimates. According to Bayes decision theory, the proposed algorithm satisfies the condition of minimum-error-rate classificatin. We apply this algorithm to NPM (Neural Prediction Model) for speech recognition, and derive new disrminative training algorithms. Experimental results on ten Korean digits recognition have shown the reduction of 37.5% of the number of recognition errors.

• Proceedings of the Korean Society for Technology of Plasticity Conference
• /
• 2004.10a
• /
• pp.75-78
• /
• 2004

A mesh generation algorithm adapted to the mesh density map using the Delaunay mesh generation technique is developed. In the finite element analyses of the forging processes, the numerical error increases as the process goes on because of discrete property of the finite elements or severe distortion of elements. Especially, in the region where stresses and strains are concentrated, the numerical discretization error will be highly increased. However, it is too time consuming to use a uniformly fine mesh in the whole domain to reduce the expected numerical error. Therefore, it is necessary to construct locally refined mesh at the region where the error is concentrated such as at the die corner. In this study, the point insertion algorithm is used and the mesh size is controlled by moving nodes to optimized positions according to a mesh density map constructed with a posteriori error estimation. An optimization technique is adopted to obtain a good position of nodes. And optimized smoothing techniques are also adopted to have smooth distribution of the mesh and improve the mesh element quality.

• Proceedings of the Computational Structural Engineering Institute Conference
• /
• 2002.10a
• /
• pp.86-93
• /
• 2002

A new error estimation method is proposed. This utilizes the variation of energy functional about the mapping function between the global and the master elements. The resultant system of equations is the weak form of the generalized conservation checks. However, This formulation has an important information about the relations between the connected elements. In other words, some relations between the connected elements are obtained and these can be used very usefully to measure it posteriori error. In this paper, the explicit formulations are presented for the 1-dimensional model and the 2-dimensional model problems. Numerical results are provided for first order shear deformation theody of beam model and the plane stress problem.

• Korean Journal of Mathematics
• /
• v.21 no.3
• /
• pp.293-304
• /
• 2013

In this paper we analyze an error estimator for the lowest-order triangular Raviart-Thomas mixed finite element which is based on solution of local problems for the error. This estimator was proposed in [Alonso, Error estimators for a mixed method, Numer. Math. 74 (1996), 385{395] and has a similar concept to that of Bank and Weiser. We show that it is asymptotically exact for the Poisson equation if the underlying triangulations are uniform and the exact solution is regular enough.

• Journal of the Korean Mathematical Society
• /
• v.52 no.5
• /
• pp.891-906
• /
• 2015

In this work we analyze a postprocessing scheme for improving the guaranteed error bound based on the equilibrated fluxes for the P1 conforming FEM. The improved error bound is shown to be asymptotically exact under suitable conditions on the triangulations and the regularity of the true solution. We also present some numerical results to illustrate the effect of the postprocessing scheme.

• Journal of Korea Multimedia Society
• /
• v.18 no.7
• /
• pp.886-894
• /
• 2015

In this paper, we conduct sensitivity analysis of parameters used for inverse modeling in order to estimate the PM₁₀ emissions from the 16 areas in East Asia accurately. Parameters used in sensitivity analysis are R, the observational error covariance matrix, and B, a priori (background) error covariance matrix. In previous studies, it was used with the predetermined parameter empirically. Such a method, however, has difficulties in estimating an accurate emissions. Therefore, an automatically determining method for the most suitable value of R and B with an error measurement criteria and posteriori emissions accuracy is required. We determined the parameters through a sensitivity analysis, and improved the accuracy of posteriori emissions estimation. Inverse modeling methods used in the emissions estimation are pseudo inverse, NNLS (Nonnegative Least Square), and BA(Bayesian Approach). Pseudo inverse has a small error, but has negative values of emissions. In order to resolve the problem, NNLS is used. It has a unrealistic emissions, too. The problems are resolved with BA(Bayesian Approach). We showed the effectiveness and the accuracy of three methods through case studies.

• Communications of the Korean Mathematical Society
• /
• v.16 no.4
• /
• pp.691-701
• /
• 2001

We computed zero mean Gaussian of average error bounds pf Simpsons quadrature with convariances in [2]. In this paper, we compute zero mean Gaussian of average error bounds between Simpsons quadrature and composite Simpsons quadra-ture on four consecutive subintervals. The reason why we compute these on subintervals is because these results enable us to compute a posteriori error bounds on the whole interval in the later paper.

• Proceedings of the KIEE Conference
• /
• 1987.07a
• /
• pp.585-588
• /
• 1987