• Korean Management Science Review
• /
• v.16 no.1
• /
• pp.1-9
• /
• 1999

The purpose of this paper is to develope efficient techniques for implementing the dual simplex method. In this paper we proposed one artificial row technique to get an initial dual feasible basic solution, a dual steepest-edge method coupled with a dropping row selection rule, and an anti-degeneracy technique which resembles the EXPAND procedure for the primal simplex method. The efficiency of the above techniques is shown by experiments. Finally, the dual simplex method is shown to be superior to the primal simplex method when it is used in the integer programming.

• Journal of Advanced Marine Engineering and Technology
• /
• v.28 no.5
• /
• pp.801-810
• /
• 2004

The Primal-Dual Interior-Point (PDIP) method is currently one of the fastest emerging topics in optimization. This method has become an effective solution algorithm for large scale nonlinear optimization problems. such as the electric Optimal Power Flow (OPF) and natural gas and electricity OPF. This study describes major theoretical developments of the PDIP method as well as practical issues related to implementation of the method. A simple quadratic problem with linear equality and inequality constraints

• Journal of Korean Society of Industrial and Systems Engineering
• /
• v.21 no.45
• /
• pp.93-100
• /
• 1998

The researches of Linear Programming are Khachiyan Method, which uses Ellipsoid Method, and Karmarkar, Affine, Path-Following and Interior Point Method which have Polynomial-Time complexity. In this study, Karmarkar Method is more quickly solved as 50 times then Simplex Method for optimal solution. but some special problem is not solved by Karmarkar Method. As a result, the algorithm by APL Language is proved time efficiency and optimal solution in the Primal-Dual interior point algorithm. Furthermore Karmarkar Method and Primal-Dual interior point Method is compared in some examples.

• Nuclear Engineering and Technology
• /
• v.53 no.7
• /
• pp.2371-2376
• /
• 2021

As the γ-ray energy increases, a reconstructed image becomes noisy and blurred due to the penetration of the γ-ray through the coded mask. Therefore, the thickness of the coded mask was increased for high energy regions, resulting in severely decreased the performance of the detection efficiency due to self-collimation by the mask. In order to overcome the limitation, a modified uniformly redundant array γ-ray imaging system using dual anti-mask method was developed, and its performance was compared and evaluated in high-energy radiation region. In the dual anti-mask method, the two shadow images, including the subtraction of background events, can simultaneously contribute to the reconstructed image. Moreover, the reconstructed images using each shadow image were integrated using a hybrid update maximum likelihood expectation maximization (h-MLEM). Using the quantitative evaluation method, the performance of the dual anti-mask method was compared with the previously developed collimation methods. As the shadow image which was subtracted the background events leads to a higher-quality reconstructed image, the reconstructed image of the dual anti-mask method showed high performance among the three collimation methods. Finally, the quantitative evaluation method proves that the performance of the dual anti-mask method was better than that of the previously reconstruction methods.

• Journal of Biomedical Engineering Research
• /
• v.13 no.1
• /
• pp.25-32
• /
• 1992

A new restoration method of chest X -ray image (dual project filter) was proposed to improve SNR(signal to noise ratio) characteristics. In this method, a priori Information of system and anatomical structure and statistics of projected object are used in the design of filter. Dual projection filter varies its parameters, adapting to the local regions of chest(lung region, mediasternum, subdiaphragm) and the structure of chest (bone, tissue, blood vessel, bronchia). The performance of Dual Projection Filter was 0.1-0.2dB better than Dual Sensor Wiener Filter, which was used for initial estimate of Dual Porjection Filter.

• Bulletin of the Korean Mathematical Society
• /
• v.51 no.4
• /
• pp.1175-1186
• /
• 2014

We construct new binary optimal self-dual codes of length 50. We develop a construction method for binary self-dual codes with a fixed-point-free automorphism of order 2. Using this method, we find new binary optimal self-dual codes of length 52. From these codes, we obtain Lee-optimal self-dual codes over the ring $\mathbb{F}_2+u\mathbb{F}_2$ of lengths 25 and 26.

• Journal of the Korean Operations Research and Management Science Society
• /
• v.24 no.1
• /
• pp.27-38
• /
• 1999

In this paper, Primal-Interior Dual-Simplex method(PIDS) and Dual-Interior Primal-Simplex method(DIPS) are developed for the general bounded linear programming. Two methods were implemented and compared with other pricing techniques for the Netlib. linear programming problems. For the PIDS, it shows superior performance to both most nagative rule and dual steepest-edge method since it practically reduces degenerate iterations and has property to reduce the problem. For the DIPS, pt requires less iterations and computational time than least reduced cost method. but it shows inferior performance to the dynamic primal steepest-edge method.

• Journal of the Korean Institute of Telematics and Electronics S
• /
• v.35S no.4
• /
• pp.20-29
• /
• 1998

In this paper, we find the values of near-end crosstalk coefficients in dual-offset stripline of a FR-4 multilayer PCB by an analytic method and a HSPICE simulation method, define calculation errors inananlytic method and the application range, simulate near-end crosstalk coefficients of the FCT(Fast CMOS TTL) in complicated dual-offset stripline by HSPICE and analyze near-end crosstalk and far-end crosstalk coefficients in dual-offset stripline. So, we analyze coupling structure of the near-end crosstalk and far-end crosstalk in the complicated dual-offset striplines that are 1[pF] capacitors termainated, and define a coupling formula of near-end crosstalk and far-end crosstalk coefficients dual-offset striplines.

• Speech Sciences
• /
• v.11 no.1
• /
• pp.43-52
• /
• 2004

This paper concerns an effective dual-channel noise reduction method to increase the performance of speech recognition in a car environment. While various single channel methods have already been developed and dual-channel methods have been studied somewhat, their effectiveness in real environments, such as in cars, has not yet been formally proven in terms of achieving acceptable performance level. Our aim is to remedy the low performance of the single and dual-channel noise reduction methods. This paper proposes an effective dual-channel noise reduction method based on a high-pass filter and front-end processing of the eigendecomposition method. We experimented with a real multi-channel car database and compared the results with respect to the microphones arrangements. From the analysis and results, we show that the enhanced eigendecomposition method combined with high-pass filter indeed significantly improve the speech recognition performance under a dual-channel environment.

• Proceedings of the Korea Electromagnetic Engineering Society Conference
• /
• 2001.11a
• /
• pp.165-168
• /
• 2001

This Paper presents the design method and performance characteristics of a dual band balun. The design method for dual balun is based on the lumped element eqiuvalent circuit of quater-wave transformaer. By employing the proposed configuration and the derived formulas, dual band balun are designed and simulated and manufactured. The proposed design method and equivalent circuit can make it easy to adapt to designing of ceramic multi-layer chip type dual band balun. The dual band will find applications in wireless communication circuits.