• 한국통신학회논문지
• /
• 제21권10호
• /
• pp.2563-2578
• /
• 1996

In this paper, channel error characteristics are investigated to alleviate the channel error propagation problem of the digital TV transmission systems. First, error propagation problems, which are mainly caused by the inter-frame dependancy and variable length coding of the MPEG-2 baseline encoder, are intensively analyzed. Next, existing channel resilient schemes are systematically classified into two kinds of schemes; one for the encoder and the other for the decoder. By comparing the performance and implementation cost, the encoder side schemes, such as error localization, layered coding, error resilience bit stream generation techniques, are described in this paper. Also, in an effort to consider the parcticality of the real transmission situation, an efficient error detection scheme for a decoder system is proposed by employing a priori information of the bit stream syntas, checking the encoding conditions at the encoder stage, and exploiting the statistics of the image itself. Finally, subsequent error concealment technique based on the DCT coefficient recovery algorithm is adopted to evaluate the performance of the proposed error resilience technique. The computer simulation results show that the quality of the received image is significantly improved when the bit error rate is as high as 10$^{-5}$ .

• 한국전산구조공학회논문집
• /
• 제13권4호
• /
• pp.485-500
• /
• 2000

본 논문에서는 element-free Galerkin(EFG) 방법에 기반한 적응적 정적균열진전해석기법을 제시하였다. 균열진전 매단계마다 적응적해석을 수행함으로써 전체 해석의 일관성과 정밀성을 동시에 확보할 수 있었다. 균열진전과정에 있어서의 적응적해석은 산정된 오차지표에 따라 적분을 위한 격자구조에 따라 절점을 추가하고 소거하는 과정을 통해 구현되었다. 이 때 사용된 오차지표는 원 EFG해석결과 얻어진 응력과 절점응력을 다시 투영한 응력의 차에 의해 얻어졌다. 제안된 해석기법의 타당성과 효용성을 수치예제에 의해 검증하였다. 그 결과 제안된 해석기법이 균열진전해석시 효율적으로 적용될 수 있음을 보였다.

• Journal of Advanced Marine Engineering and Technology
• /
• 제21권1호
• /
• pp.66-70
• /
• 1997

In this paper a new learning algorithm for curvature smoothing and improved generalization for multi-layer neural networks is proposed. To enhance the generalization ability a constraint term of hidden neuron activations is added to the conventional output error, which gives the curvature smoothing characteristics to multi-layer neural networks. When the total cost consisted of the output error and hidden error is minimized by gradient-descent methods, the additional descent term gives not only the Hebbian learning but also the synaptic weight decay. Therefore it incorporates error back-propagation, Hebbian, and weight decay, and additional computational requirements to the standard error back-propagation is negligible. From the computer simulation of the time series prediction with Santafe competition data it is shown that the proposed learning algorithm gives much better generalization performance.

• 센서학회지
• /
• 제23권1호
• /
• pp.1-6
• /
• 2014

In many medical image devices, dc noise often prevents normal diagnosis. In wireless capsule endoscopy systems, multipath fading through indoor wireless links induces inter-symbol interference (ISI) and indoor electric devices generate impulsive noise in the received signal. Moreover, dc noise, ISI, and impulsive noise are also found in optical fiber communication that can be used in remote medical diagnosis. In this paper, a blind signal processing method based on the biased probability density functions of constant modulus error that is robust to those problems that can cause error propagation in decision feedback (DF) methods is presented. Based on this property of robustness to error propagation, a DF version of the method is proposed. In the simulation for the impulse response of optical fiber channels having slowly varying dc noise and impulsive noise, the proposed DF method yields a performance enhancement of approximately 10 dB in mean squared error over its linear counterpart.

• 대한전기학회논문지:전력기술부문A
• /
• 제52권10호
• /
• pp.571-576
• /
• 2003

XML Web Service based on internet can cause problems on transmission speed and data error. Also system analysis results simulated by several different research groups can hardly have reliability because of error data that come from improperly managed files. In order to solve this problems, algorithm sever using XML Web Service is shared on the internet so widely that various application programs based on basic analysis module with a united IO can be developed. And also XML Dataset DB is interacted with XML Web Service, which prevents propagation of error data. It causes to improve reliabilityon the load flow analysis result and solve the problems on data error or transmission speed that can possibly come from internet.

• 한국음향학회지
• /
• 제20권6호
• /
• pp.75-81
• /
• 2001

본 논문에서는 근거리 효과를 이용하여 음원의 위치를 추정하는 탐지 기법을 해양환경에 적용할 경우에 발생할 수 있는 위치추정 오차에 관하여 분석하였다. 삼각 (triangulation) 알고리듬과 파면곡률 (wavefront curvature) 알고리듬 등을 이용하는 근거리 탐지 기법은 음파가 2차원 평면 (방위, 거리)에서 전달된다고 가정한다. 그러나 해양환경은 2차원 평면이 아닌 3차원 공간 (방위, 거리, 수심)이므로 음파전달에 따른 오차가 발생할 수 있다. 3차원 공간을 가정한 경우에도 해양에서의 다중경로 음파전달을 고려하지 않았다면 역시 오차가 발생하게 될것이다. 근거리 탐지 기법의 위치추정 오차를 분석하기 위하여 다중경로 음파전달모델과 파면곡률을 이용한 초점 빔형성 (focused beamforming)기법을 이용하여 시뮬레이션하였다. 분석결과 수중음속구조, 해저면 수심, 해저면 경사와 음원의 거리 등에 따라 위치추정 오차가 달라짐을 볼 수 있었다.

• 한국지능시스템학회논문지
• /
• 제5권3호
• /
• pp.83-91
• /
• 1995

본 연구에서는 종래에 학습법으로 널리 이용되고 있는 역전파학습법의 문제점으로 지적되어 온 학습에 많은 시간이 걸리는 점과 국소적 최적해에 해가 수렴하여 오차가 충분히 작게 되지 않는 등의 문제점을 해결하기 위해, Hu에 의해 고안된 random tabu 탐색법을 이용하여 신경회로망의 연결강도를 최적화하는 학습알고리즘을 새로이 제안하였다. 그리고 이 방법을 배타적 논리합 문제에 적용하여 기존의 역전파학습법과 학습상수 $, $에 tabu탐색법을 이용한 결과와 비교 검토하여 본 방법이 국소적 최적해에 수렴하지 않고 수렴정도를 개선할 수 있음을 확인하였다.

• 대한전기학회:학술대회논문집
• /
• 대한전기학회 1991년도 하계학술대회 논문집
• /
• pp.896-899
• /
• 1991

This paper presents the neural network method to predict the Korea composition stock price index. The error back-propagation method is used to train the multi-layer perceptron network. Ten of the various economic indices of the past 7 Nears are used as train data and the monthly transition of the composition stock price index is represented by five output neurons. Test results of this method using the data of the last 18 months are very encouraging.

• 한국정보과학회:학술대회논문집
• /
• 한국정보과학회 1998년도 가을 학술발표논문집 Vol.25 No.2 (2)
• /
• pp.327-329
• /
• 1998

Up-Propagation is an algorithm for inverting and learning neural network generative models. Sensory input is processed by inverting a model that generates patterns from hidden variables using top-down connections. The inversion process is iterative, utilizing a negative feedback loop that depends on an error signal propagated by bottom-up connections. The error signal is also used to learn the generative model from examples. the algorithm is benchmarked against principal component analysis in experiments on images of handwritten digits.

• ETRI Journal
• /
• 제17권1호
• /
• pp.11-22
• /
• 1995

This paper proposes a modified error function to improve the error back-propagation (EBP) algorithm for multi-Layer perceptrons (MLPs) which suffers from slow learning speed. It can also suppress over-specialization for training patterns that occurs in an algorithm based on a cross-entropy cost function which markedly reduces learning time. In the similar way as the cross-entropy function, our new function accelerates the learning speed of the EBP algorithm by allowing the output node of the MLP to generate a strong error signal when the output node is far from the desired value. Moreover, it prevents the overspecialization of learning for training patterns by letting the output node, whose value is close to the desired value, generate a weak error signal. In a simulation study to classify handwritten digits in the CEDAR [1] database, the proposed method attained 100% correct classification for the training patterns after only 50 sweeps of learning, while the original EBP attained only 98.8% after 500 sweeps. Also, our method shows mean-squared error of 0.627 for the test patterns, which is superior to the error 0.667 in the cross-entropy method. These results demonstrate that our new method excels others in learning speed as well as in generalization.