• 전기학회논문지
• /
• 제56권7호
• /
• pp.1331-1338
• /
• 2007

A new method for detecting and eliminating the Electrocardiogram(ECG) artifact from the scalp Electroencephalogram(EEG) is proposed. Based on the single channel EEG, the proposed method consists of 4 procedures: emphasizing the R-wave of ECG artifact from EEG using the least squares acceleration(LSA) filter, detecting the R-wave from the LSA filtered EEG using the phase space method and R-R interval, generating the delayed impulse synchronized to the R-wave and elimination of the ECG artifacts based on the adaptive digital filter using the impulse and raw EEG. The performance of the proposed method was evaluated in the two separating parts of R-wave detection and, ECG estimation and elimination from EEG. In the R-wave detection, the proposed method showed the mean error rate of 6.285(%). In the ECG estimation and elimination using simulated and/or real EEG recordings, we found that the ECG artifacts were successfully estimated and eliminated in comparison with the conventional multi-channel techniques, in which independent component analysis and ensemble average method are used. From this we can conclude that the proposed method is useful for the detecting and eliminating the ECG artifact from single channel EEG and simple for ambulatory/portable EEG monitoring system.

• Wind and Structures
• /
• 제27권3호
• /
• pp.199-211
• /
• 2018

This paper deals with wind fragility and risk analysis of high rise buildings subjected to stochastic wind load. Conventionally, such problems are dealt in full Monte Carlo Simulation framework, which requires extensive computational time. Thus, to make the procedure computationally efficient, application of metamodelling technique in fragility analysis is explored in the present study. Since, accuracy by the conventional Least Squares Method (LSM) based metamodelling is often challenged, an efficient Moving Least Squares Method based adaptive metamodelling technique is proposed for wind fragility analysis. In doing so, artificial time history of wind load is generated by three wind field models: i.e., a simple one based on alongwind component of wind speed; a more detailed one considering coherence and wind directionality effect, and a third one considering nonstationary effect of mean wind. The results show that the proposed approach is more accurate than the conventional LSM based metamodelling approach when compared to full simulation approach as reference. At the same time, the proposed approach drastically reduces computational time in comparison to the full simulation approach. The results by the three wind field models are compared. The importance of non-linear structural analysis in fragility evaluation has been also demonstrated.

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 2005년도 ICCAS
• /
• pp.2120-2125
• /
• 2005

The principal idea of vibration isolation is to filter out the response of the system over the corner frequency. The isolation objectives are to transmit the attitude control torque within the bandwidth of the attitude control system and to filter all the high frequency components coming from vibration equipment above the bandwidth. However, when a reaction wheels or control momentum gyros control spacecraft attitude, vibration inevitably occurs and degrades the performance of sensitive devices. Therefore, vibration should be controlled or isolated for missions such as Earth observing, broadcasting and telecommunication between antenna and ground stations. For space applications, technicians designing controller have to consider a periodic vibration and disturbance to ensure system performance and robustness completing various missions. In general, past research isolating vibration commonly used 6 degree order freedom isolators such as Stewart and Mallock platforms. In this study, the vibration isolation device has 3 degree order freedom, one translational and two rotational motions. The origin of the coordinate is located at the center-of-gravity of the upper plane. In this paper, adaptive notch filter finds the disturbance frequency and the reference signal in filtered-x least mean square is generated by the notch frequency. The design parameters of the notch filter are updated continuously using recursive least square algorithm. Therefore, the adaptive filtered-x least mean square algorithm is applied to the vibration suppressing experiment without reference sensor. This paper shows the experimental results of an active vibration control using an adaptive filtered-x least mean squares algorithm.

• Journal of the Korean Wood Science and Technology
• /
• 제45권4호
• /
• pp.399-408
• /
• 2017

Research on the rapid and accurate prediction of physical properties of wood using near-infrared (NIR) spectroscopy has attracted recent attention. In this study, partial least squares analysis was performed between NIR spectra and air-dried specific gravity of five domestic conifer species including larch (Larix kaempferi), Korean pine (Pinus koraiensis), red pine (Pinus densiflora), cedar (Cryptomeria japonica), and cypress (Chamaecyparis obtusa). Fifty different lumbers per species were purchased from the five National Forestry Cooperative Federations of Korea. The air-dried specific gravity of 100 knot- and defect-free specimens of each species was determined by NIR spectroscopy in the range of 680-2500 nm. Spectral data preprocessing including standard normal variate, detrend and forward first derivative (gap size = 8, smoothing = 8) were applied to all the NIR spectra of the specimens. Partial least squares analysis including cross-validation (five groups) was performed with the air-dried specific gravity and NIR spectra. When the performance of the regression model was expressed as $R^2$ (coefficient of determination) and root mean square error of calibration (RMSEC), $R^2$ and RMSEC were 0.63 and 0.027 for larch, 0.68 and 0.033 for Korean pine, 0.62 and 0.033 for red pine, 0.76 and 0.022 for cedar, and 0.79 and 0.027 for cypress, respectively. For the calibration model, which contained all species in this study, the $R^2$ was 0.75 and the RMSEC was 0.37.

• 응용통계연구
• /
• 제34권2호
• /
• pp.239-254
• /
• 2021

이 논문에서는, 2개의 혼합된 t-분포(TP-T)의 오차과정을 따르는 이질적 자기회귀 (HAR) 모형을 이용하여, 한국 코로나 (COVID-19) 확진자 수 데이터에 대한 시계열 분석, 즉 추정과 예측에 대하여 연구한다. HAR-TP-T 시계열 모형을 고려하여 HAR 모형의 계수 뿐 아니라 TP-T 오차과정의 모수를 추정하고자 단계별 추정법을 제안한다. 본 연구에서 제안하고 있는 단계별 추정법은, HAR 계수 추정을 위해서는 통상적 최소제곱추정법을 채택하고, TP-T 모수 추정을 위해서는 최대우도추정법을 이용한다. 단계별 추정법에 대한 모의실험을 수행하여, 성능이 우수함을 입증한다. 한국 코로나 확진자 수에 대한 실증적 데이터 분석에서, HAR 모형에서의 차수 p = 2, 3, 4에 대해, 모형의 평균제곱오차가 최소가 되도록 하는 최적화 시간간격(optimal lag)을 포함하여, 여러가지 시간간격을 고려한 HAR-TP-T 모형의 모수 추정값을 계산한다. 제안된 단계별 추정방법과 기존의 MLE만의 방법을, 추정 결과를 제시함으로 함께 비교한다. 본 연구에서 제안하고 있는 추정은 두 가지의 오차 측면, 즉 HAR 모형의 평균제곱오차와 잔차분포에 대한 밀도함수 추정의 평균제곱오차, 두 측면에서 모두 우수함을 입증하였다. 나아가, 추정 결과를 활용한 코로나 확진자 수 예측을 수행하였고, 예측정확도의 한 측도로서 mean absolute percentage error (MAPE)를 계산하여 0.0953%의 매우 작은 오차값을 얻었다. 본 연구에서 선택한 최적화 시간간격을 고려한 HAR-TP-T 시계열 모형 및 단계별 추정 방법은, 정확한 한국 코로나 확진자 수 예측 성능을 제공한다고 할 수 있다.

• 한국통신학회논문지
• /
• 제29권2C호
• /
• pp.272-282
• /
• 2004

본 논문에서는 정규화 혼합 norm을 이용한 다중 채널 영상 복원 기법을 제안한다. 채널 영상 간 및 채널 영상내의 결정론적 정보를 이용한 다중 채널 영상 복원에 관한 문제를 고려한다. 제안 방식에서는 각 채널 영상에 대해 LMS (Least Mean Square) 및 LMF (Least Mean Fourth) 및 완화 함수를 결합시킨 부가 함수가 제안된다. 더불어, LMS 및 LMF의 상대적 기여도를 제한하기 위한 혼합 norm 매개 변수 및 완화 함수의 중요성을 제어하는 정규화 매개 변수는 반복 영상으로부터 예측된 각 채널의 노이즈 분포에 의해 결정되어 진다. 제안된 방식은 각 채널 영상의 첨부 노이즈 형태에 대한 사전 정보 없이 복원 과정이 가능하다는 점과 두 매개 변수를 반복 과정에서 부분적으로 복원된 영상으로부터 조절할 수 있는 특성을 갖고 있다.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제4권1호
• /
• pp.45-61
• /
• 2010

In the communication channel, the received signal is affected by many factors that can cause errors. These effects mean that received signal strength (RSS) based methods incur more errors in measuring distance and consequently result in low precision in the location detection process. As one of the approaches to overcome these problems, we propose using direction calibration to improve the performance of the RSS-based method for distance measurement, and sequentially a weighted least squares (WLS) method using reliability factors in conjunction with a conventional RSS weighting matrix is proposed to solve an over-determined localization process. The proposed scheme focuses on the features of the RSS method to improve the performance, and these effects are proved by the simulation results.

• Journal of the Korean Statistical Society
• /
• 제23권1호
• /
• pp.167-178
• /
• 1994

To estimate coefficient matrix in autoregressive model, usually ordinary least squares estimator or unconditional maximum likelihood estimator is used. It is unknown that for univariate AR(p) model, unconditional maximum likelihood estimator gives better power property that ordinary least squares estimator in testing for unit root with mean estimated. When autoregressive model contains multiple unit roots and unconditional likelihood function is used to estimate coefficient matrix, the seperation of nonstationary part and stationary part of the eigen-values in the estimated coefficient matrix in the limit is developed. This asymptotic property may give an idea to test for multiple unit roots.

• 대한의용생체공학회:의공학회지
• /
• 제26권2호
• /
• pp.87-93
• /
• 2005

A new method for real-time estimating left ventricular relaxation time constant (T) from the left ventricular (LV) pressure waveform, based on the isovolumic relaxation model, is proposed. The presented method uses a recursive least squares (RLS) algorithm to accomplish real-time estimation. A new criterion to detect the end-point of the isovolumic relaxation period (IRP) for the estimation of T is also introduced, which is based on the pattern analysis of mean square errors between the original and reconstructed pressure waveforms. We have verified the performance of the new method in over 4,600 beats obtained from 70 patients. The results demonstrate that the proposed method provides more stable and reliable estimation of τ than the conventional 'off-line' methods.

• 응용통계연구
• /
• 제25권3호
• /
• pp.531-541
• /
• 2012

In this study, we consider various methods to estimate the weights of a pairwise comparison matrix in the Analytic Hierarchy Process widely applied in various decision-making fields. This paper uses a data dependent simulation to evaluate the statistical accuracy, minimum violation and minimum norm of the obtaining weight methods from a reciprocal symmetric matrix. No method dominates others in all criteria. Least squares methods perform best in point of mean squared errors; however, the eigenvectors method has an advantage in the minimum norm.