• Proceedings of the Acoustical Society of Korea Conference
• /
• spring
• /
• pp.85-88
• /
• 2004

칼만/위너 필터 같은 기존의 음성향상 알고리즘은 잡음의 선험적 지식을 요구하고, 음성신호와 추정신호의 오차분산을 최소화하는데 중점을 두었다. 따라서, 잡음에 대한 통계적 추정에 오류가 있을 경우 결과에 악영향을 미칠 수 있다. 그러나 $H_\infty$ 필터는 잡음에 대한 어떠한 가정이나 선험적 지식을 요구하지 않는다. $H_\infty$ 필터는 최소상계(Upper Bound Least)를 적용하여 추정된 모든 신호들로부터 최소 에러 신호를 갖는 최상의 추정신호를 찾아내므로 칼만/위너 필터보다 잡음의 변화에 강인하다. 본 논문에서는 학습 신호로부터 은닉 마코프 모델의 파리미터를 추정한 후, 오염된 신호를 고정된 개수의 $H_\infty$ 필터를 통과시켜 각 출력에 가중된 합으로 향상된 음성 신호를 구한다. 음성의 통계적 특성을 이용하여 모델 파라미터를 추정하는 은닉 마코프 모델과 잡음의 변화에 강인한 $H_\infty$ 알고리즘을 사용해서, 다중 $H_\infty$필터에 의한 강인한 음성향상 방법을 제안하였다.

• Journal of the Institute of Electronics Engineers of Korea SC
• /
• v.40 no.4
• /
• pp.251-263
• /
• 2003

This paper presents matrix inequality conditions for Η$_2$control and Η$_2$controller design method of linear time-invariant descriptor systems with parameter uncertainties in continuous and discrete time cases, respectively. First, the necessary and sufficient condition for Η$_2$control and Η$_2$ controller design method are expressed in terms of LMI(linear matrix inequality) with no equality constraints in continuous time case. Next, the sufficient condition for Hi control and Η$_2$controller design method are proposed by matrix inequality approach in discrete time case. Based on these conditions, we develop the robust Η$_2$controller design method for parameter uncertain descriptor systems and give a numerical example in each case.

• Transactions of the Korean Society of Mechanical Engineers A
• /
• v.29 no.1 s.232
• /
• pp.74-80
• /
• 2005

A number of creep rupture data for type 316LN stainless steels were collected through literature survey or experimental data produced in KAERI. Using these data, polynomial equations for predicting creep life were obtained by Larson-Miller (L-M), Orr-Sherby-Dorn (O-S-D) and Manson-Haferd (M-H) parameters using time-temperature parametric (TTP) methods. Standard error of estimate (SEE) values for the each parameter was obtained with different temperatures through the statistical process of the creep data. The results of L-M, O-S-D and M-H methods showed good creep-life prediction, but M-H method showed better agreement than L-M and O-S-D methods. Especially, it was found that SEE values of M-H method at $700^{\circ}C$ were lower than that of L-M and O-S-D methods.

• The Journal of the Korea institute of electronic communication sciences
• /
• v.9 no.10
• /
• pp.1207-1214
• /
• 2014

The first frame of a GOP is encoded in intra mode which generates a larger number of bits. In addition, the first frame is used for the inter mode encoding of the following frames. Thus the encoding results of the intra frame affects the first frame as well as the following frames. Traditionally, the quantization parameter for an intra frame is determined only depending on the bpp not considering the characteristics of the intra frame. For accurate intra frame encoding, we should consider not only bpp but also the complexity of the video sequence and the output bandwidth. In this paper, we propose a real-time quantization model which is used to calculate the quantization parameter for an intra frame encoding based on the investigation on the characteristics of a GOP. It is shown by experimental results that the proposed quantization model captures the characteristics of an intra frame effectively and the proposed method for model parameters accurately estimates the real values.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.18 no.5
• /
• pp.518-523
• /
• 2017

This paper proposes a novel denoising method based on non-local(NL) means. The NL-means algorithm is effective for removing an additive Gaussian noise, but the denoising parameter should be controlled depending on the noise level for proper noise elimination. Therefore, the proposed method optimizes the denoising parameter according to the noise levels. The proposed method consists of two processes: off-line and on-line. In the off-line process, the relations between the noise level and the denoising parameter of the NL-means filter are analyzed. For a given noise level, the various denoising parameters are applied to the NL-means algorithm, and then the qualities of resulting images are quantified using a structural similarity index(SSIM). The parameter with the highest SSIM is chosen as the optimal denoising parameter for the given noise level. In the on-line process, we estimate the noise level for a given noisy image and select the optimal denoising parameter according to the estimated noise level. Finally, NL-means filtering is performed using the selected denoising parameter. As shown in the experimental results, the proposed method accurately estimated the noise level and effectively eliminated noise for various noise levels. The accuracy of noise estimation is 90.0% and the highest Peak Signal-to-noise ratio(PSNR), SSIM value.

• The Korean Journal of Pesticide Science
• /
• v.7 no.1
• /
• pp.58-65
• /
• 2003

Various physicochemical parameters for the active ingredients of 245 herbicides were calculated to develope a diagnosis and estimation system for utility as herbicide. The range of physico-chemical parameters for each inhibitors of photo system II (H1), acetolactate synthase (ALS) (H2) and herbicides were confirmed. The distribution ranges of 85% dependence for each physicochemical parameters were Obs.logP :$-0.90\sim4.50$, dipol moment: $1.80\sim12.22$ (debye), molecular refractivity: $53.0\sim104.0(cm^3/mol)$, polarizability: $19.0\sim37.0(\AA^3)$, HOMO energy: $-9.98\sim-7.34$ (eV), LUMO energy:$-2.76\sim0.40$ (eV), Van der Waals molecular volumes: $558.0\sim995.0(cm^3)$, molecular weight: $202.0\sim430.0$ (amu) and surface areas (Grids): $194.0\sim356.0(\AA^2)$, hydration energy: $-10.16\sim114.7$ Kcal/mol, respectively. It is suggested that MR and polarizability constants will be able to distinguish between herbicides and medicinal drugs. Results revealed that various compounds based on the range of physicochemical parameters of herbicides could be diagnosed and estimated.

• 제어로봇시스템학회:학술대회논문집
• /
• 2000.10a
• /
• pp.23-23
• /
• 2000

In this paper, we proposed the problems of robust stability and 개bust H$_{\infty}$ control of discrete time-delay linear st.stems with Frobenius norm-bounded uncertainties. The existence condition and the design method of robust H$_{\infty}$ state feedback control]or are given. Through some changes of variables and Schur complement, the obtained sufficient condition can be rewritten as an LMI(linear matrix inequality) form in terms of all variables.

• Proceedings of the KIEE Conference
• /
• 2001.07b
• /
• pp.608-610
• /
• 2001

In this paper, 6/4 Switched Reluctance Motor(SRM) which has simple structure and little switching element is selected basic analysis model. In order to reduce torque ripple causing noise and vibration, we execute optimization of geometric parameters (stator and rotor pole arc) and electrical parameters (turn-on angle and turn-of angle) by means of combining Fletcher-Reeves's Conjugate Directions and Finite Element Method (FEM) considering driving circuits. When considering the switching condition according to inductance profile, torque characteristics is influenced by geometric and electrical parameters importantly. The pole arc and switching angle of the optimum can also obtain the low torque ripple without high currents.

• Journal of the Institute of Convergence Signal Processing
• /
• v.3 no.2
• /
• pp.76-82
• /
• 2002

The objective of this paper is to design a robust controller for a flexible robot arm manipulator using LMI(Linear Matrix Inequality) theory, and confirm its effectiveness through experimentation. We first describe a modeling Process of the flexible arm in order to get a mathematical model, and then discuss how to approximately obtain the uncertainty of the model for robust control. As to the control system design, we adopt the LMI-based H$_{\infty}$ synthesis algorithm which has the merits of eliminating the regularity restrictions attached to the Riccati-based methods. As a result of this, we can cope with the parameter variation (that is, modeling uncertainty) due to the tip-load variation. Finally we confirm the effectiveness of the controller through experiment and simulation.

• Proceedings of the KIEE Conference
• /
• 1998.11b
• /
• pp.454-456
• /
• 1998

Traditional adaptive control algorithms are not robust to dynamic uncertainties. The adaptive control algorithms developed previously to deal with dynamic uncertainties do not facilitate quantitative design. We proposed a new robust adaptive control algorithms consists of an $H^{\infty}$ suboptimal control law and a robust parameter estimator. Numerical examples showing the effectiveness of the $H^{\infty}$ adaptive scheme are provided.