• Journal of the Korean Institute of Telematics and Electronics
• /
• v.22 no.4
• /
• pp.24-32
• /
• 1985

This paper presents the design method for multivalued logic functions using $I^2L$ circuits. First, the a비orithm that transforms delta functions into discrete functions of a truncated difference is obtained. The realization of multivalued logic circuits by this algorithm is discussed. And then, the design method is achieved by mixing discrete functions and delta functions using the modified algorithm for given multivalued truth tables. The techniques discussed here are easily extended to multi-input and multi-output logic circuits.

• Proceedings of the IEEK Conference
• /
• 2003.11a
• /
• pp.7-10
• /
• 2003

This paper proposes the multiple objects focusing algorithm. Given multiple objects at different distances from a camera, we assume that one object is well-focused and the others are out-of-focused. The proposed auto-focusing algorithm is summarized as follows: (i) detects edges from an input image, (ⅱ) estimates the radius of PSF (Point Spread Function) across the edge, (ⅲ) gather edge points having same radius of PSF, (ⅳ) segments the image into regions with the same radius of PSF, and (ⅴ) restores the each segmented region using the corresponding PSF.

• Proceedings of the IEEK Conference
• /
• 2000.06e
• /
• pp.81-84
• /
• 2000

This paper proposes a new sigma delta modulator structure based on a LMS(Least Mean Square) algorithm that minimizes the quantization noise. The proposed architecture provides 40dB SNR improvement and 35dB wider dynamic range over conventional sigma delta modulation. The proposed architecture provides superior performance especially when the input signal is small.

• Journal of the Korean Institute of Telematics and Electronics S
• /
• v.35S no.1
• /
• pp.96-104
• /
• 1998

A new gray-scale thinning algorithm using local min/max operations is proposed. In this method, erosion and dilation properties of local min/max operations are using for generating new rides and detecting ridges in gray scale image, and gray-scale skeletons are gradually obtained by accumulating the detected ridges. This method can be applicable to the unsegmented image in which object are not specified, and the obtained skeletons correspond to the ridges (high gray values) of an input image.

• The Journal of the Acoustical Society of Korea
• /
• v.15 no.3E
• /
• pp.19-26
• /
• 1996

In this paper, convergence properties of the LMS, LMF, and LVCMS algorithms are investigated under the assumption of the uncorrelated Gaussian input data. By treating these algorithms as special cases of more general algorithm family, unified results on these algorithms are obtained. First the upper bound on the step size parameter is obtained. Second, an expression for misadjustment is obtained. These theoretical results confirm earlier LMS works. Further, the results explain why the LMS and LVCMS algorithms are experiencing difficulties with plant noise having heavier tailed densities. Simulation results agree with theoretical expectation closely for various plant noise statistics.

• Proceedings of the KIEE Conference
• /
• 1988.07a
• /
• pp.305-308
• /
• 1988

Pole assignment controller with variable forgetting factor is generalizaed to allow the output and/or input variance to be reduced. The algorithm can give significant reductions in variance for little extra computational effort and is presented for servo-tracking using leat-squares estimation. Moreover, the use of a variable forgetting factor with correct choice of information bound can avoid 'blowing-up' of the covariance matrix of the estimates and subsequent unstable control.

• Proceedings of the Korea Society for Industrial Systems Conference
• /
• 1996.10a
• /
• pp.211-216
• /
• 1996

This paper propose a fuzzy regression method using fuzzy neural networks when a membership value is attached to each input-output pair. First, an architecture of fuzzy nerual networks with fuzzy weights and fuzzy biases is shown. Next a cost function is defined using the fuzzy output from the fuzzy neural network and the corresponding target output with a membership value.A learning algorithm is derived from the cost function. The derived learning algorithm trains the fuzzy neural network so that the level set of the fuzzy output includes the target output. Last, the proposed method is applied to the quality evaluation problem of injection molding.

• Proceedings of the IEEK Conference
• /
• 2002.06a
• /
• pp.347-350
• /
• 2002

In the input buffer switch system, VOQ(Virtual Output Queue) archives 100% throughput. The VOQ with the systolic architecture maintains an uniform performance regardless of a number of Packet class and output port, so that it doesn't have a limitation of scalability. In spite of these advantages, the systolic architecture VOQ is difficult to change sorting order In this paper, we Proposed a systolic architecture VOQ which support weighted round robin(WRR) algorithm to provide with flow control service.

• ETRI Journal
• /
• v.28 no.1
• /
• pp.59-66
• /
• 2006

Learning behaviors of a radial basis function network (RBFN) using a singular value decomposition (SVD) and stochastic gradient (SG) algorithm, together named RBF-SVD-SG, for odor sensing systems are analyzed, and a fast training method is proposed. RBF input data is from a conducting polymer sensor array. It is revealed in this paper that the SG algorithm for the fine-tuning of centers and widths still shows ill-behaving learning results when a sufficiently small convergence coefficient is not used. Since the tuning of centers in RBFN plays a dominant role in the performance of RBFN odor sensing systems, our analysis is focused on the center-gradient variance of the RBFN-SVD-SG algorithm. We found analytically that the steadystate weight fluctuation and large values of a convergence coefficient can lead to an increase in variance of the center-gradient estimate. Based on this analysis, we propose to use the least mean square algorithm instead of SVD in adjusting the weight for stable steady-state weight behavior. Experimental results of the proposed algorithm have shown faster learning speed and better classification performance.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.14 no.3
• /
• pp.559-564
• /
• 2010

The GMDH(Group Method of Data Handling) algorithm can be used to predict the complex nonlinear systems. The traditional GMDH algorithm produces the prdicted output of the system model in the output layer through the input layer and the intermediate layers as the prescribed process. The outputs of each layer are produced only by the outputs of the former layer. However, in the traditional GMDH algorithm, though the optimal structure of each layer is derived, the overall structure may not be derived optimally. To overcome this problem, GMDH prediction model which has the overall optimal structure is constructed by feeding back the error between the predicted output and the real output. This can make the prediction more precise. The capability improvement of the proposed algorithm compared to the traditional algorithm is verified through computer simulation.