• Journal of applied mathematics & informatics
• /
• 제8권1호
• /
• pp.231-242
• /
• 2001

We showed in [2] that if r≤2, zero mean Gaussian of average error of the Trapezoidal rule is proportional to h/sub i//sup 2r+3/ on the interval [0,1]. In this paper, if r≥3, we show that zero mean Gaussian of average error of the Trapezoidal rule is bounded by Ch⁴/sub i/h⁴/sub j/.

• 품질경영학회지
• /
• 제30권1호
• /
• pp.118-132
• /
• 2002

This paper presents a process incapability index to provide manager with various information of process and to reduce cost. The introduced process incapability indices indicate information about mean and variance of manufacturing process and variance of inspection process to evaluate process capability using ratio of variance and difference between target and mean to specification. This model can be used by the scale of six sigma management.

• 한국운동역학회지
• /
• 제15권4호
• /
• pp.25-42
• /
• 2005

The study was to investigate kinematic difference between flexed and extended type of knee during Jigeo-Cha-Gi in Taekwon-Do. For this subjects participated were consisted of weights of fin (1), bantam (1) and welter class(1) of male 3 national representative level skilled in two type(flexed and extended type) of Jigeo-Cha-Gi. 3D cinematography analysis was performed for motion analysis and Kwon 3D ver. 3.1) was used for 3D coordinates & analysis variables calculation. In Temporal variable there was no significant difference statistically in all phases & total elapsed time between flexed and extended type, but flexed type was delayed more 0.016 sec than extended type. In displacement of COG there was significant difference in level of p<.05 showing longer mean 6.13 cm in case of flexed type than extended type in displacement of COG during all phase and too significant difference in level of p<.01 showing longer mean 4.4 cm in case of flexed type than extended type in displacement of COG in follow through phase. In velocity of COG there was significant difference in level of p<.001 showing higher mean 15.53cm/s in case of flexed type than extended type in velocity of COG(Y direction) during targeting phase and peak velocity(Y) was more fast 8.74 cm/s in extended type than flexed type. In velocity of leading leg in forward direction(Y) there was significant difference in level of p<.05 showing higher thigh mean value in case of flexed type than extended type but showing higher foot mean value in extended type at level of p<.001 than flexed type in velocity of COG(Y direction). In velocity of leading leg in vertical direction(Z) there was no significant difference in the second & third phase in case of vertical velocity level, but momentum transferred efficiently form proximal to distal endpoint. In front-back & right-left orientation angle of trunk there was possibility of more stable Jigeo-Cha-Gi in extended than flexed type by decreasing in right-left orientation angle of trunk. In relative angle of lower leg(hip, knee, ankle) there was significant difference in level of p<.001 showing longer mean 32.74 deg. in case of flexed type than extended type in hip joint during the second phase but maintained insufficient extended knee of mean 168 deg. in targeting phase.

• 한국통신학회논문지
• /
• 제26권12A호
• /
• pp.2043-2049
• /
• 2001

In this paper we study the convergence behavior of the least mean fourth(LMF) algorithm where the error raised to the power of four is minimized for a multiple sinusoidal input and Gaussian measurement noise. Here we newly obtain the convergence equation for the sum of the mean of the squared weight errors, which indicates that the transient behavior can differ depending on the relative sizes of the Gaussian noise and the convergence constant. It should be noted that no similar results can be expected from the previous analysis by Walach add Widrow.

• The Journal of the Acoustical Society of Korea
• /
• 제18권2E호
• /
• pp.3-9
• /
• 1999

In this paper we study the convergence behavior of the least mean fourth(LMF) algorithm where the error raised to the power of four is minimized for a multiple sinusoidal input and Gaussian measurement noise. Here we newly obtain the convergence equation for the sum of the mean of the squared weight errors, which indicates that the transient behavior can differ depending on the relative sizes of the Gaussian noise and the convergence constant. It should be noted that no similar results can be expected from the previous analysis by Walach and Widrow/sup [1]/.

• 대한전기학회논문지:시스템및제어부문D
• /
• 제54권1호
• /
• pp.40-47
• /
• 2005

This paper concerns on Filtered-x least mean square (FXLMS) algorithm for adaptive estimation of feedforward control parameters. The conditions for convergence in ensemble mean of the FXLMS algorithm are derived and the directional convergence properties are discussed from a new geometric vector analysis. The convergence and its directionality are verified along with some computer simulations.

• 대한전자공학회:학술대회논문집
• /
• 대한전자공학회 2001년도 제14회 신호처리 합동 학술대회 논문집
• /
• pp.915-918
• /
• 2001

In this paper we study a new convergence behavior of the least mean fourth (LMF) algorithm where the error raised to the power of four is minimized for a multiple sinusoidal input and Gaussian measurement noise. Here we newly obtain the convergence equation for the sum of the mean of the squared weight errors, which indicates that the transient behavior can differ depending on the relative sizes of the Gaussian noise and the convergence constant. It should be noted that no similar results can be expected from the previous analysis by Walach and Widrow.

• 대한전자공학회논문지
• /
• 제25권6호
• /
• pp.672-679
• /
• 1988

In this paper, attempts have been made to code images employing a separating mean vector quantizer(SMVQ). Then we analyzed the performance of the SMVQ experimentally as well as analytically. The results of simulation with natural images are presented. But, conclusively the performance of the SMVQ technique is not better than that of the conventional vector quantizer. In this paper, a brief analysis in which we revealed that the performance, based on the mean square error measure, of the SMVQ is not favorable is favorable is discussed.

• 대한수학회논문집
• /
• 제16권4호
• /
• pp.691-701
• /
• 2001

We computed zero mean Gaussian of average error bounds pf Simpsons quadrature with convariances in [2]. In this paper, we compute zero mean Gaussian of average error bounds between Simpsons quadrature and composite Simpsons quadra-ture on four consecutive subintervals. The reason why we compute these on subintervals is because these results enable us to compute a posteriori error bounds on the whole interval in the later paper.

• 산업경영시스템학회지
• /
• 제14권24호
• /
• pp.31-41
• /
• 1991

The producer's to tolerance concept (in contrast to the consumer's tolerence) fellows naturally from the notion of continuous loss function Emerging producer tolerance methodology has proceeded without benefit of carefully defined assumptions. We rectifg this omission and consider the process mean deviation. Then develope the models for which determine the tolerenes in cases of the-nominal-the-best, the-smaller-the-better and the-larger-the-better. Finally the numerical examples are Presented to illustrated these results and also the sensitivity analysis to apprehend the changes of loss tolerances varying with mean is presented.