• 한국정밀공학회지
• /
• 제24권4호
• /
• pp.76-83
• /
• 2007

A new neural network classifier is proposed for the automatic real-time surface inspection of high-speed cold steel strips having 11 different types of defects. 46 geometrical and gray-level features are extracted for the defect classification. 3241 samples of Posco's Kwangyang steel factory are used for training and testing the neural network classifier. The developed classifier produces plausible 15% error rate which is much better than 20-30% error rate of human vision inspection adopted in most of domestic steel factories.

• 전자공학회논문지A
• /
• 제30A권11호
• /
• pp.29-39
• /
• 1993

For an economical or timely reason, the types of subscriber loops for ISDN basic rate access service will, at least initially, be that of the pre-existing telephone network, that is, 2-wire twisted-pair metallic cable. Therefore, it is important to assess the high-speed digital transmission capability of the 2-wire subscriber loops. In this paper, the impairments such as intersymbol interference, crosstalk, and impulse niose are modeled. And then, symbol error probabilities in terms of SNR in DSL system using ECH and 2B1Q line code are analytically derived and numerically calculated. Also, the relationship between the quality of service required in ISDN subscriber section and the loop parameters is investigated.

• 한국생산제조학회지
• /
• 제18권6호
• /
• pp.681-689
• /
• 2009

Disk grinding was often applied to deburring process in order to enhance the final product quality. Inherent chamfering capability of the flexible disk grinding process in the early stage was analyzed with respect to various process parameters including workpiece length, wheel speed, depth of cut and feed. Initial chamfered edge defined as arc zone was characterized with local radius of curvature. Averaged radius and arc zone ratio was well evaluated using neural network system. Additional neural network analysis adding workpiece length showed enhance performance in predicting arc zone ratio and curvature radius with reduced error rate. A process condition design parameter was estimated using remaining input and output parameters with the prediction error rate lower than 2.0% depending on the relevant input parameter combination and neural network structure composition.

• 한국공작기계학회논문집
• /
• 제17권5호
• /
• pp.123-130
• /
• 2008

In order to clarify detailed mechanism of the flexible disk grinding system, workpiece length was introduced and its performance was evaluated. Flat zone ratio increased as the workpiece length increased. Increasing wheel speed and depth of cut also enhanced process performance by producing larger flat zone ratio. Neural network system was successfully applied to predict minimum depth of engagement and flat zone ratio. An additional input parameter as workpiece length to the neural network system enhanced the prediction performance by reducing error rate. By rearranging the Input combinations to the network, the workpiece length was precisely predicted with the prediction error rate lower than 2.8% depending on the network structure.

• ETRI Journal
• /
• 제27권5호
• /
• pp.525-532
• /
• 2005

In this paper, we propose and present implementation results of a high-speed turbo decoding algorithm. The latency caused by (de)interleaving and iterative decoding in a conventional maximum a posteriori turbo decoder can be dramatically reduced with the proposed design. The source of the latency reduction is from the combination of the radix-4, center to top, parallel decoding, and early-stop algorithms. This reduced latency enables the use of the turbo decoder as a forward error correction scheme in real-time wireless communication services. The proposed scheme results in a slight degradation in bit error rate performance for large block sizes because the effective interleaver size in a radix-4 implementation is reduced to half, relative to the conventional method. To prove the latency reduction, we implemented the proposed scheme on a field-programmable gate array and compared its decoding speed with that of a conventional decoder. The results show an improvement of at least five fold for a single iteration of turbo decoding.

• 한국지능시스템학회논문지
• /
• 제5권2호
• /
• pp.3-12
• /
• 1995

A controller utilizing fuzzy logic is developed to control the speed of a motor in a washing machine by choosing an appropriate phase. Due to the hardship imposed on obtaining a result from a relation established for inputs, present speed and present rate of speed, and ouput, a phase, of the system that can be tested against an experimental result, it is impossible to apply a genetic algorithm to fine-tune the fuzzy logic controller. To avoid this difficulty, a proper assumption that the parameters of an if-part of a primary fuzzy logic controller have a functional relationship with an error between computed values and experimental ones in made. Setting up of a fuzzy relationship between the parameters and the errors is then achieved through experimentally obtained data. Genetic Algorithm is then applied to this secondary fuzzy logic controller to verify the fuzzy logic. In the verification process, the primary fuzzy logic controller is used in obtaining experimental results. In this way the kind of difficulty in obtaining enough experimental values used to verify the fuzzy logic with genetic algorithm is gotten around. Selection of the parameters that would produce the least error when using the secondary fuzzy logic controller is done with applying genetic algorithm to the then-part of the controller. In doing so the optimal values for the parameters of the if-part of the primary fuzzy logic controller are assumed to be contained. The experimental result presented in the paper validates the assumption.

• 로봇학회논문지
• /
• 제18권2호
• /
• pp.182-188
• /
• 2023

Due to the acceleration of an aging society, the need for lower limb exoskeletons to assist gait is increasing. And for use in daily life, it is essential to have technology that can accurately estimate gait phase even in the walking environment and walking speed of the wearer that changes frequently. In this paper, we implement an LSTM-based gait phase estimation learning model by collecting gait data according to changes in gait speed in outdoor level ground and stair environments. In addition, the results of the gait phase estimation error for each walking environment were compared after learning for both max hip extension (MHE) and max hip flexion (MHF), which are ground truth criteria in gait phase divided in previous studies. As a result, the average error rate of all walking environments using MHF reference data and MHE reference data was 2.97% and 4.36%, respectively, and the result of using MHF reference data was 1.39% lower than the result of using MHE reference data.

• 산업경영시스템학회지
• /
• 제27권2호
• /
• pp.52-60
• /
• 2004

High-speed machining is one of the most effective technologies to improve productivity. Because of the high speed and high feed rate, high-speed machining can give great advantages for the machining of dies and molds. High speed machine tool makers try to find best machining condition with the one that they have built. Machine builders need to develop test specimen since it helps finding characteristics o( machine tools when the machining properties of the specimen are analyzed. This paper develops test specimen to identify features of the main spindle, the feeding device, and the frame of a machine tool.

• 전자공학회논문지B
• /
• 제33B권11호
• /
• pp.159-166
• /
• 1996

In this paper, the method for improving the speed of convergence and learning rate of back propagation algorithms is proposed which update the learning rate parameter and momentum term for each weight by generated error, changely the output layer of neural network generates a high value in the case that output value is far from the desired values, and genrates a low value in the opposite case this method decreases the iteration number and is able to learning effectively. The effectiveness of proposed method is verified through the simulation of X-OR and 3-parity problem.

• 전자공학회논문지
• /
• 제53권8호
• /
• pp.145-149
• /
• 2016

수중음파통신은 지상과는 다른 환경적 문제에 의해 고속 데이터 전송률을 얻기 위해서는 여러 가지 어려움이 있다. 고속 데이터 전송을 위해서 다중안테나 전송방식에 대해서 많은 연구가 진행되고 있다. V-BLAST 알고리즘은 무선통신에서 다중안테나 전송방식에 적용되고 검파방식이다. 본 논문에서는 수중음파채널에 대해서 여 MIMO 시스템의 VBLAST 검파 방식에 대해서 성능을 분석한다.