• 제어로봇시스템학회:학술대회논문집
• /
• 1991.10a
• /
• pp.768-772
• /
• 1991

The Self Creating and organizing (SCO) is a new architecture and one of the unsupervized learning algorithm for the artificial neural network. SCO begins with only one output node which has a sufficiently wide response range, and the response ranges of all the nodes decrease with time. Self Creating and Organizing Neural Network (SCONN) decides automatically whether adapting the weights of existing node or creating a new node. It is compared to the Kohonen's Self Organizing Feature Map (SOFM). The results show that SCONN has lots of advantages over other competitive learning architecture.

• Proceedings of the IEEK Conference
• /
• 1999.06a
• /
• pp.460-463
• /
• 1999

In this paper, we proposed an block-based evolvable neural network(BENN). The BENN can optimize it＇s structure and weights simultaneously. It can be easily implemented by FPGA whose connection and internal functionality can be reconfigured. To solve the local minima problem that is caused gradient descent learning algorithm, genetic algorithms are applied for optimizing the proposed evolvable neural network model.

• Proceedings of the IEEK Conference
• /
• 1999.06a
• /
• pp.94-97
• /
• 1999

In this paper, a new real-time neural network connection admission controller is proposed. The proposed controller measures traffic flows, cell loss rate and cell delay periodically each classes. The Neural network learns the relation between those measured information and service quality by real-time. Also the proposed controller uses the DWRR multiplexer with buffer dedicated to every traffic source in order to measure the delay that cells experience in buffer. Experimental result shows that the proposed method can control effectively heterogeneous traffic sources with diverse QoS requirement.

• Journal of the Korean Institute of Telematics and Electronics C
• /
• v.35C no.5
• /
• pp.59-68
• /
• 1998

This paper proposes a neural network algorithm for a channel assignment in cellular mobile communications. The proposed algorithm is developed base on hopfield neural network in order to minimize the number of channel without a confliction between cells. To compare the performance of the proposed algorithm, we used seven benchmark problems selected from kunz's and funabiki's papers. Experimental results show that the convergence times are reduced form 27% to 66% compared with Kunz's and funabiki's algorithm and vonvergence rates are improved to 100%.

• Journal of the Korean Institute of Telematics and Electronics B
• /
• v.32B no.12
• /
• pp.1663-1669
• /
• 1995

This paper is a study on the recognition of Korean proverb using neural network and Markov model. The neural network uses, at the stage of training neurons, features such as the rate of zero crossing, short-term energy and PLP-Cepstrum, covering a time of 300ms long. Markov models were generated by the recognized phoneme strings. The recognition of words and proverbs using Markov models have been carried out. Experimental results show that phoneme and word recognition rates are 81. 2%, 94.0% respectively for Korean proverb recognition experiments.

• Proceedings of the Optical Society of Korea Conference
• /
• 1991.06a
• /
• pp.23-28
• /
• 1991

A single-layer neural network with 4$\times$4 input neurons and 4 output neurons is optically implemented. Holographic lenslet arrays are used for the e optical interconnection topology, a liquid crystal light valve(LCLV) is used for controlling optical interconection weights. Using a Perceptron learning rule, it classifics input patterns into 4 different categories. It is shown that the performance of the adaptive neural network depends on the learning rate, the correlation of input patterns, and the nonlinear characteristic properties of the liquid crystal light valve.

• Journal of the Korean Data and Information Science Society
• /
• v.24 no.3
• /
• pp.659-667
• /
• 2013

If the regression model or the propensity model is correct, the unbiasedness of the estimator using doubly robust imputation can be guaranteed. Using a neural network instead of a logistic regression model for the propensity model, the estimators using doubly robust imputation are approximately unbiased even though both assumed models fail. We also propose a doubly robust estimator of ratio form using population information of an auxiliary variable. We prove some properties of proposed theory by restricted simulations.

• 제어로봇시스템학회:학술대회논문집
• /
• 1990.10a
• /
• pp.339-343
• /
• 1990

This paper describes a neural-network-based methodology for providing a potential solution in the area of process fault diagnosis. The existing neural network for fault diagnosis learn fault node by using pairs of single-symptom-single-cause only. But in real plants, the effect of a fault propagates continuously from it's origin; different sensor values reflect this. In this paper, we suggest a new method which can handle the effect of symptom propagation. The proposed method can find the exact origin of the fault of which the symptom is propagated continuously with time.

• Proceedings of the Korean Institute of IIIuminating and Electrical Installation Engineers Conference
• /
• 1999.11a
• /
• pp.132-135
• /
• 1999

In this work, two approaches were proposed for the recognition of partial discharge patterns. The first approach was neural network with backpropagation algorithm, and the second approach was angle calculation between two operator vectors. PD signal were detected using three electrode systems; IEC(b), needle-plane and CIGRE method II electrode system. Both of neural network and angle comparison method showed good recognition performance for the patte군 similar to the trained patterns. And the number of operators to be used had a great influence on the recognition performance to the untrained patterns.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.26 no.8B
• /
• pp.1078-1084
• /
• 2001

In this paper, fuzzy-ARTMAP neural network is applied for compensating the nonlinearity of satellite communication channel. The fuzzy-ARTMAP is made of using fuzzy logic and ART neural network. By a match tracking process with vigilance parameter, fuzzy ARTMAP neural network achieves a minimax learning rule that minimizes predictive error and maximizes generalization. Thus, the system automatically learns a minimal number of recognition categories, or hidden units, to meet accuracy criteria. Simulation studies are performed over satellite nonlinear channels. The performance of proposed fuzzy-ARTMAP equalizer is compared with MLP-basis equalizers.