• The Journal of the Korea Contents Association
• /
• v.15 no.4
• /
• pp.1-8
• /
• 2015

This paper presents a Electrocardiogram(ECG) identification method using adaptive weight based on Least Mean Square Error(LMSE) optimization. With a preprocessing for noise suppression, we extracts the average ECG signal and its standard deviation at every time instant. Then the extracted information is stored in database. ECG identification is achieved by matching an input ECG signal with the information in database. In computing the matching scores, the standard deviation is used. The scores are computed by applying adaptive weights to the values of the input signal over all time instants. The adaptive weight consists of two terms. The first term is the inverse of the standard deviation of an input signal. The second term is the proportional one to the standard deviation between user SAECGs stored in the DB. Experimental results show up to 100% recognition rate for 32 registered people.

• Journal of Biosystems Engineering
• /
• v.33 no.6
• /
• pp.438-445
• /
• 2008

The purpose of this study was to develop a measurement model based on PLS (Partial least square) method for blood pressures. Measurement system for blood pressure signals consisted of pressure sensor, va interface and embedded module. A mercury sphygmomanometer was connected with the measurement system through 3-way stopcock and used as reference of blood pressures. The blood pressure signals of 20 subjects were measured and tests were repeated 5 times per each subject. Total of 100 data were divided into a calibration set and a prediction set. The PLS models were developed to determine the systolic and the diastolic blood pressures. The PLS models were evaluated by the standard methods of the British Hypertension Society (BHS) protocol and the American Association for the Advancement of Medical Instrumentation (AAMI). The results of the PLS models were compared with those of MAA (maximum amplitude algorithm). The measured blood pressures with PLS method were highly correlated to those with a mercury sphygmomanometer in the systolic ($R^2=0.85$) and the diastolic blood pressure ($R^2=0.84$). The results showed that the PLS models were the effective tools for blood pressure measurements with high accuracy, and satisfied the standards of the BHS protocol and the AAMI.

• Journal of Korean Society of Occupational and Environmental Hygiene
• /
• v.17 no.1
• /
• pp.21-30
• /
• 2007

To establish the Fourier-transform infra-red spectrophotometry (FTIR) direct-on-filter(DOF) technique as a useful analytical method for quartz in respirable dust samples, influence of the interference should be corrected. This study was designed to compare three methods of correction for cristobalite when quantifying the content of quartz, including the least square, the optimum choice and the spectral subtraction methods. Respirable dust, created in a dust chamber containing the standard material of quartz, cristobalite was collected using a cyclone equipped with a 25 mm, $0.8{\mu}m$ pore size DM filter as a collection medium. The quartz weights overestimated about 100% when mixed of cristobalite by measure using 799 cm-1 absorption peak of quartz. The quartz weights appeared over estimated by optimum choice, spectral subtraction and least square method in mixtures of 33% cristobalite were 90.3%, 60.1% and about 4.3%, respectively. The least square method have been adopted to correction methods of cristobalite and satisfactory results have been obtained. The results of this study suggest that, when correcting for effect of cristobalite on quantitative analysis of quartz in respirable dust by FTIR direct-on-filter method, the least square method produce the most unbiased results compared with those of other correction methods.

• Journal of Industrial Technology
• /
• v.25 no.B
• /
• pp.157-164
• /
• 2005

Two methods, the bin averaging method and least square method, are often used in calibrating wind turbine test sites. The objective of this work was to determine a better method to predict the wind speed at wind turbine installing point. The calibration was done at the test site on a complex terrain located in Daegwallyeong, Korea. It was performed for two different cases based on the IEC 61400-12 power performance measurement standard. The wind speeds averaged for 10 minutes ranged between 4 m/s and 16 m/s. The wind-direction bins of each meteorological mast were 10 degrees apart, and only the bins having data measured for more than 24 hours were employed for the test site calibration. For both cases, the two methods were found to yield almost same results which estimated real wind speed very closely.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2007.06a
• /
• pp.406-409
• /
• 2007

This paper proposes an algorithm that predicts current electrostatic charge in a factory. The algorithm based on LSM(Least Square Method) dynamically takes the number of sample while calculating the value of electrostatic charge. The simulation results show that the proposed algorithm gains 73.18161 standard deviation with 95% trust probability and is better than conventional algorithm. We design the electrostatic prediction embedded system based on pxa255 with the proposes algorithm.

• Analytical Science and Technology
• /
• v.16 no.3
• /
• pp.185-190
• /
• 2003

We applied modified least square method (MLS) and ordinary least square method (OLS) to 1st order equation for the comparison of the uncertainties calculated by these methods. The uncertainty calculated by OLS covered statistically safe interval because it was over-estimated in many cases of measurement and concentration level. But, if the uncertainty of the concentration as a reference value was comparably large (about 5% of the relative standard deviation of random scattering from the regression line and about 7% of relative standard uncertainty of reference values), then uncertainty calculated by OLS was seriously under-estimated at high concentration level. It was revealed that the calculated uncertainty didn't cover statistically safe interval at the stated confidence level. It was found that the method, MLS, described in the previously article would be valid for this calculation of uncertainty.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.22 no.4
• /
• pp.399-404
• /
• 2012

Evolutionary algorithms such as genetic algorithm (GA) have been proven their effectiveness when applying to the design of fuzzy models. However, it tends to suffer from computationally expensWive due to the slow convergence speed. In this study, we propose an approach to develop fuzzy models by means of an improved differential evolution (IDE) to overcome this limitation. The improved differential evolution (IDE) is realized by means of an orthogonal approach and differential evolution. With the invoking orthogonal method, the IDE can search the solution space more efficiently. In the design of fuzzy models, we concern two mechanisms, namely structure identification and parameter estimation. The structure identification is supported by the IDE and C-Means while the parameter estimation is realized via IDE and a standard least square error method. Experimental studies demonstrate that the proposed model leads to improved performance. The proposed model is also contrasted with the quality of some fuzzy models already reported in the literature.

• Journal of the Korean Society for Precision Engineering
• /
• v.15 no.7
• /
• pp.167-174
• /
• 1998

This paper presents an unoccluded cylindrical object pose measurement using a slit beam laser in which a robot recognizes all of the unoccluded objects from the top of jumbled objects, and picks them up one by one. The elliptical equation parameters of a projected curve edge on a slice are calculated using LSM. The coefficients of standard elliptical equation are compared with these parameters to estimate the object pose. The hamming distances between the estimated coordinates and the calculated ones are extracted as measures to evaluate a local constraint and a smoothing surface curvature. The edges between slices are linked using error function based on the edge types and the hamming distances. The linked edges on slices are compared with the model object's length to recognize the unoccluded object. This proposed method may provide a solution to the automation of part handling in manufacturing environments such as punch press operation or part assembly.

• Journal of the Institute of Convergence Signal Processing
• /
• v.12 no.3
• /
• pp.163-168
• /
• 2011

In this paper, we proposed non-linear gamma curve algorithm for gamma correction. The previous non-linear gamma curve algorithm is generated by the least square polynomial using the Gauss-Jordan inverse matrix. However, the previous algorithm has some weak points. When calculating coefficients using inverse matrix of higher degree, occurred truncation errors. Also, only if input sample points are existed regular interval on 10-bit scale, the least square polynomial is accurately works. To compensate weak-points, we calculated accurate coefficients of polynomial using eigenvalue and orthogonal value of mat11x from singular value decomposition (SVD) and QR decomposition of vandemond matrix. Also, we used input data part segmentation, then we performed polynomial curve fitting and merged curve fitting results. When compared the previous method and proposed method using the mean square error (MSE) and the standard deviation (STD), the proposed segmented polynomial curve fitting is highly accuracy that MSE under the least significant bit (LSB) error range is approximately $10^{-9}$ and STD is about $10^{-5}$.

• Journal of Korean Society of Industrial and Systems Engineering
• /
• v.16 no.28
• /
• pp.137-143
• /
• 1993

Generally there are three methods to derive an approximation formula: 1) approching standard normal distribution by appropriate changing variable 2) using standardization variable for expansion 3) deriving approximation formula by direct method. This paper present correction terms having the form of $C_{1/v^{n/2}}/{\;}+{\;}C_2{\;}(n=1,2)$ with respect to $x^2_{\alpha}(v)$ distribution (${\nu}{\;}{\leq}{\;}30$), which minimize the error by EDA method and least square method.