• Proceedings of the Acoustical Society of Korea Conference
• /
• spring
• /
• pp.77-80
• /
• 2004

인간은 "습관적 피치 레벨" 즉 자연스럽게 말할 때 평균적으로 사용하는 피치를 갖는다. 하지만 음성에 잡음이 첨가 되면 이 피치가 불규칙하게 바뀌게 된다. 이점을 이용하여 음성의 잡음레벨을 측정할 수 있다. 본 논문에서는 입력음성의 에너지를 구하고 일정 에너지레벨 이상에서의 구간에 대해 NAMDF(Normalized Average Magnitude Difference Function)방법으로 피치를 구하고, 각 프레임을 피치단위로 분절한 뒤 인근 피치간의 유사도를 측정하여 입력음성데이터의 잡음레벨을 검출하는 방법을 제안하였다.

• Proceedings of the Acoustical Society of Korea Conference
• /
• autumn
• /
• pp.73-76
• /
• 2004

인간은 "습관적 피치 레벨" 즉 자연스럽게 말할 때 평균적으로 사용하는 피치를 갖는다. 하지만 음성에 잡음이 첨가되면 이 피치가 불규칙하게 바뀌게 된다. 이점을 이용하여 음성의 잡음레벨을 측정할 수 있다. 본 논문에서는 입력음성의 에너지를 구하고 일정 에너지레벨 이상에서의 구간에 대해 NAMDF(Normalized Average Magnitude Difference Function)방법으로 피치를 구하고, 각 프레임을 피치단위로 분절한 뒤 인근 피치간의 유사도를 측정하여 입력음성데이터의 잡음레벨을 검출하는 방법을 제안하였다.

• Journal of Institute of Control, Robotics and Systems
• /
• v.12 no.3
• /
• pp.300-306
• /
• 2006

In this paper an error analysis of 3-dimensional GNSS attitude determination system is given. The attitude error covariance matrix is derived and analyzed. It implies that attitude errors are affected by the baseline length and configuration, the satellites numbers and geometry, receiver measurement noises and the nominal attitude of the vehicle. By defining Euler Angle Dilution Of Precision (EADOP) which is analogous to GDOP, roll, pitch and yaw errors can be efficiently analyzed. However the expression of the attitude error is too complex to get some intuitions. Therefore with a commonly adopted assumption, new expressions for attitude error are derived. The formulas are easy to compute and represent the attitude error as a function of the nominal attitude of a vehicle, the baseline configuration and the receiver noise. Using the formula, the accuracy of the attitude can be analytically predicted without the computer simulations. Applications to some widely used configurations reveal the effectiveness of the proposed method.

• Journal of Institute of Control, Robotics and Systems
• /
• v.9 no.2
• /
• pp.107-116
• /
• 2003

This paper presents a systematic design methodology for fault tolerant controller against a fault in actuators and sensors of linear stochastic systems with uncertainties. The scheme is based on fault detection and diagnosis(isolation and estimation) using a bank of robust two-stage Kalman filters, and accommodation of the actuator fault by eigenstructure assignment and immediate compensation of the sensor's faulty measurement. In order to clarify the fault feature in test statistics of residual, noise reduction method is given by multi-scale discrete wavelet transform. The effectiveness of our approach Is shown via simulations for a VTOL(vertical take-off and landing) aircraft subjected to parameter variations, external disturbances, process and sensor noises.

• Journal of the Korea Computer Graphics Society
• /
• v.4 no.1
• /
• pp.21-29
• /
• 1998

This paper deals with character animation using motion capture. The captured motion requires the editing process to smooth the jerky motion by the sensor noise, or to combine several clip-motion libraries. For this purpose, we describe a simple technique for editing the captured motion using the Kalman filter technique. Our formulation allows the generated motion to satisfy the kinematic constraints of the human model. Furthermore, it provides us with a multi-level control mechanism of the motion resolution by changing the uncertainty of the measurement model and the seamless motion transition.

• Proceedings of the KIEE Conference
• /
• 2008.09a
• /
• pp.214-217
• /
• 2008

To detect partial discharge of 154kV joint box, we have made experiment by using the HFCT sensor. Generally the signals which are detected in partial discharge test of underground power transmission cable are accompanied with both noises of high voltage and noises of surrounding Power cable. The most noise in near to end part of joint box is corona, beside other noises flowed from surrounding area. Partial discharge test is difficulty due to these noises. First, we test reliability on both injection of calibration signal in NJB and removal of low frequency. After that, we had analyzed frequencies by measuring signals in IJB with 300[m] distance from NJB. Also we had measured S/N ratio by using the indirected injection method of calibration signal in IJB. In this experiment, two measurement methods were difference of detection acquisition, but these had the equal frequency properties.

• Proceedings of the KIEE Conference
• /
• 2008.09a
• /
• pp.218-221
• /
• 2008

Due to a variable noises in the field underground lines, clear technology of noises is very important in the diagnosis technology of partial discharge. Therefore we progressed filter test to improve the measuring test of partial discharge in a field. We applied to field test after making W-LC filter for partial discharge measurement, and then designed experimental environment such as under- ground distribution lines of field. Measuring sensor for test was UHF sensor. And we analysed those using the Lemke Probe and oscilloscope (TDS-3054, Tektro- nix). Finally, we obtained excellent data of noise decrement in the joint box of 22.9[kV] distribution lines from above experiment.

• Journal of the Korea Institute of Military Science and Technology
• /
• v.10 no.4
• /
• pp.145-151
• /
• 2007

This paper considers new filter gains for the ${\alpha}\;-\;{\beta}$ tracker which is optimized particularly to minimize the tracking gate size. Optimizing the performance index which is composed of tracking errors due to target maneuver and measurement noise is not different from the existing method to obtain the ${\alpha}\;-\;{\beta}$ gains. However, holding the probability 0.997 that a target exists in the tracking gate and minimizing the gate size produce the new result not similar to the existing ${\alpha}\;-\;{\beta}$ gains.

• Journal of Power Electronics
• /
• v.16 no.4
• /
• pp.1387-1395
• /
• 2016

A method for the identification of mechanical parameters based on integral calculation is presented. Both the moment of inertia and the friction constant are identified by the method developed here, which is based on well-known mechanical differential equations. The mechanical system under test is excited according to a pre-determined low-frequency sinusoidal motion, minimizing the distortion, and increasing the accuracy of the results. The parameters are identified using integral calculation, increasing the robustness of the results against measurement noise. Experimental data are supported by simulation, confirming the effectiveness of the proposed technique. The performance improvements shown here are of use in the design of speed and position controllers and observers. Owing to its simplicity, this method can be readily applied to commercial inverter products.

• Journal of Power System Engineering
• /
• v.4 no.3
• /
• pp.40-47
• /
• 2000

The turbulent flow with wake, reattachment and recirculation flow is very important from the viewpoint of engineering. But that is still difficult because of especially the unsteady problems which are related with the vehicle dynamics and the aerodynamics noise. This paper evaluate LES that can analyze about all fluid flow region including the laminar, transition and turbulent. So we compare the results of LES with those of PIV measurement and Reynolds averaging models. In conclusion, LES predicts flow behavior better than Reynolds averaging models.