• Proceedings of the Korean Society for Noise and Vibration Engineering Conference
• /
• 2000.06a
• /
• pp.1661-1666
• /
• 2000

Nowadays, the two microphone method is accepted as the standard as specified in ASTM E1050-90 for measuring in-duct acoustic properties. However, research results on using the least square method with multiple measurement points and broadband excitation have been reported for enhancing the frequency response of the two microphone method. In this paper, the effects of varying the relative measurement positions on errors in the estimation of the acoustic quantities is studied for the multiple microphone method. Both of the theoretical and experimental results show that, among every possible sensor positioning configurations, the equidistant positioning of sensors yields the smallest error within the effective measurement frequency range. In addition, it is noted that the measurement accuracy can be increased and the effective frequency range can be widened by increasing the number of equidistant sensors. Measurement examples are shown and the results support the findings.

• Journal of the Korean Society for Precision Engineering
• /
• v.19 no.4
• /
• pp.87-92
• /
• 2002

By using the signal error test, model structure of total antilock braking system consisting of electromagnetic system and hydraulic system is determined as 9th order system. For determining parameters of the ABS, using time discrete model of parametric method, parameters in time discrete model are searched by least square method. By bilinear transform, we have found the model of ABS in s domain. Afterward, experimental output data is compared with simulated output data by MATLAB haying identified parameter. As the result, experimental data is agreed with simulated data very well.

• Journal of Institute of Control, Robotics and Systems
• /
• v.15 no.1
• /
• pp.72-81
• /
• 2009

This paper presents a robust adaptive control method using model reference control strategy against autonomous robot systems with random friction nature. We approximate a nonlinear robot system model by means of a feedback linearization approach to derive nominal control law. We construct a Least Square (LS) based observer to estimate friction dynamics online and then represent a perturbed system model with respect to approximation error between an actual friction and its estimation. Model reference based control design is achieved to implement an auxiliary control in order for reducing control error in practice due to system perturbation. Additionally, we conduct theoretical study to demonstrate stability of the perturbed system model through Lyapunov theory. Numerical simulation is carried out for evaluating the proposed control methodology and demonstrating its superiority by comparing it to a traditional nominal control method.

• Proceedings of the IEEK Conference
• /
• 2002.06e
• /
• pp.133-136
• /
• 2002

EEG(Electroencephalography) is generated by electrical activity between neurons in cortical. Waveform of EEG is changed according to body and mental states. Therefore EEG is used to diagnosis of encephalophyma and epilepsy, etc. Also EEG is used to HCI(Human-Computer Interface). This paper describes estimation of orientation and location of dipole sources. The forward model is three-layer spherical head model and current dipole model. Using analytical solution, EEG is generated. Using MNLS(Minimum-Norm Least-Square) method, orientation and location of dipole moment is estimated.

• Proceedings of the IEEK Conference
• /
• 2003.07e
• /
• pp.2200-2203
• /
• 2003

In an active noise control(ANC) system using the Filtered-X least mean square(LMS) algorithm, the online secondary path modeling method by exploiting a random noise generator is applied. This method is suitable for secondary path modeling. However, it is increased the residual error of the ANC system. In this paper, we presents an ANC system improved online secondary path modeling method which is modified Kuo and Zhang model that is the secondary path estimation by the additive noise. In addition, our proposed model is used that additive noise is transformed into the signal multiplied reference signal by gain control parameter and delayed.

• Proceedings of the KIEE Conference
• /
• 2015.07a
• /
• pp.1343-1344
• /
• 2015

본 연구에서는 증분형 FCM(Incremental Fuzzy C-Means: Incremental FCM) 클러스터링 알고리즘을 기반으로 방사형 기저함수 신경회로망(Radial Basis Function Neural Networks: RBFNN) 패턴 분류기를 설계한다. 방사형 기저함수 신경회로망은 조건부에서 가우시안 함수 또는 FCM을 사용하여 적합도를 구하였지만, 제안된 분류기에서는 빅 데이터간의 적합도를 구하기 위해 증분형 FCM을 사용한다. 또한, 빅 데이터를 학습하기 위해 결론부에서 재귀최소자승법(Recursive Least Square Estimation: RLSE)을 사용하여 다항식 계수를 추정한다. 마지막으로 추론부에서는 증분형 FCM에서 구한 적합도와 재귀최소자승법으로 구한 다항식을 이용하여 최종 출력을 구한다.

• Proceedings of the KIEE Conference
• /
• 2015.07a
• /
• pp.1355-1356
• /
• 2015

본 논문에서는 Convolution기법 기반 RBFNNs 패턴 분류기를 사용한 2차원 얼굴인식 시스템을 설계한다. 제안된 방법은 특징 추출과 차원축소를 하는 컨볼루션 계층과 부분추출 계층을 교대로 연결하여 2차원 이미지를 1차원의 특징 배열로 만든다. 그 후, 만들어진 1차원의 특징 배열을 RBFNNs 패턴 분류기의 입력으로 사용하여 인식을 수행한다. RBFNNs의 조건부에는 FCM 클러스터링 알고리즘을 사용하며 연결가중치는 1차 선형식을 사용하였다. 또한 최소 자승법(LSE : Least Square Estimation)을 사용하여 다항식의 계수를 추정하였다. 제안된 모델의 성능을 평가하기 위해 CMU PIE Database를 사용한다.

• Journal of Advanced Marine Engineering and Technology
• /
• v.28 no.2
• /
• pp.228-236
• /
• 2004

In this paper. we present a neuro-fuzzy controller which unifies both fuzzy logic and multi-layered feed forward neural networks. Fuzzy logic provides a means for converting linguistic control knowledge into control actions. On the other hand. feed forward neural networks provide salient features. such as learning and parallelism. In the proposed neuro-fuzzy controller. the parameters of membership functions in the antecedent part of fuzzy inference rules are identified by using the error back propagation algorithm as a learning rule. while the coefficients of the linear combination of input variables in the consequent part are determined by using the least square estimation method. Finally. the effectiveness of the proposed controller is verified through computer simulation for an inverted pole system.

• Transactions of the Korean Society of Mechanical Engineers A
• /
• v.24 no.9 s.180
• /
• pp.2353-2360
• /
• 2000

Statistical fatigue properties of metallic materials are increasingly required for reliability design purpose. In this study, static and fatigue tests were conducted and the normal, log-normal, two -parameter Weibull distributions at the 5% significance level are compared using the Kolmogorov-Smirnov goodness-of-fit test. Parameter estimation were compared with experimental results using the maximum likelihood method and least square method. It is found that two-parameter Weibull distribution and maximum likelihood method provide a good fit for static and fatigue life data. Therefore, it is applicable to the static and fatigue life analysis of the spheroidal graphite cast iron. The P-S-N curves were evaluated using log-normal distribution, which showed fatigue life behavior very well.

• Journal of the Korean Institute of Navigation
• /
• v.13 no.2
• /
• pp.1-21
• /
• 1989

The computer can be used to display a continuously updated list or plot of vessel position. The computer that accept input data from a number of different navigation systems, e.g., Loran , Satnav, Radar, Decca, Compass, Sextant with electrical output etc., can compute the position of a vessel relative to prerecorded objects. The celestial navigation system requires the computer to do not much calculation. Calculation are for trigonometeric, linear systems, finding roots of nonlinear equation and least square estimation etc, . In order to computerize the celestrial navigation system, these calculations must be programmed. The purpose of this thesis is to study the formulation, the design and the test of calculations of the coordinates of celestial bodies, the altitude correction and the solution of the navigational triangle processes.