• The Transactions of the Korea Information Processing Society
• /
• v.4 no.2
• /
• pp.456-464
• /
• 1997

We describe a random prcess model that prvides sets of patterms whth prcisely contrlolled within-class varia-bility and between-class distinctions.We used these pattems in a simulation study wity the back-propagation netwoek to chracterize its perfotmance as we varied the process-controlling parameters,the statistical differences between the processes,and the random noise on the patterns.Our results indicated that grneralized statistical difference between the processes genrating the patterns provided a good predictor of the difficulty of the clssi-fication problem. Also we analyzed the performance of the Bayes classifier whith the maximum-likeihood cri-terion and we compared the performance of the neural network to that of the Bayes classifier.We found that the performance of neural network was intermediate between that of the simulated and theoretical Bayes classifier.

• Journal of Information Technology Application
• /
• v.3 no.4
• /
• pp.119-134
• /
• 2001

Improving software product quality is a key to increasing user satisfaction and to achieving competitive edge. There are two approaches to assure high software product quality; development process-oriented and product-oriented. There have been many efforts for improving software quality through process certification, for example, CMM, ISO 9000 family, ISO/IEC 12207, SPICE and Bootstrap. However, a good process alone cannot guarantee good product quality. A need for the evaluation of software product quality by an independent third party is growing rapidly for several reasons. We are concerned with an application of Statistical Quality Control (SQC) to the evaluation of software product quality to obtain the efficiency of evaluation processes and the objectivity of evaluation results. Methods for selecting test cases using a random sampling approach have been discussed and methods for selecting acceptance criteria with respect to software product quality have also been suggested.

• Journal of the Korean Society of Surveying, Geodesy, Photogrammetry and Cartography
• /
• v.33 no.2
• /
• pp.123-129
• /
• 2015

The Kalman filter has been widely used in the kinematic GPS positioning due to its flexibility and efficiency in computational points of view. At the same time, the relative positioning technique also provided the high precision positioning results by removing the systematic errors in the measurements significantly. However, the positioning quality may be degraded following to longer in baseline length. For this case, it is required that the remaining atmospheric effects, such as double-difference ionospheric delay and zenith wet delay, should be properly modeled by examining the characteristics of the stochastic processes. In general, atmospheric effects are estimated with the assumption of random walk, or the first-order Gauss-Markov stochastic process, which requires the precise modeling on the corresponding process noises. Therefore, we determined and provided the parameters for modelling the process noises for atmospheric effects. The auto-correlation functions are empirically determined at first, and then the parameters are extracted from the empirical auto-correlation function. In fact, the test results can be either applied directly, or used as guidance values for the modeling of process noises in the kinematic GPS positioning.

• Proceedings of the Korean Institute of Industrial Safety Conference
• /
• 2001.11a
• /
• pp.25-30
• /
• 2001

엔진이나 콤프레셔 등과 같은 기기 진동의 동특성은 불규칙적으로 변동하는 성분을 갖는 랜덤 프로세스로 그것을 수학적으로 명확히 기술하는 것은 어렵다/sup 1)/. 하지만 통계학적인 입장에서 시계열 데이터를 보면 시계열 데이터가 확률변수로서 각각의 모집단에 속한다. 따라서 이 점에 주목하여 시계열의 확률적인 특징을 추출하는 것으로, 각각의 시계열 데이터를 확률공간으로 구별하는 것이 가능하다면 시계열 데이터에 의해 표현되어진 상태의 식별가능하다는 이론이 성립된다/sup 2)/.(중략)

• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.38 no.4
• /
• pp.1-1
• /
• 2001

본 논문에서는 웨이블릿 변환을 멀티스케일 매치 필터의 관점에서 해석하고, 이를 위하여 마르코프 랜덤 필드에 묻혀있는 가우시안 형태의 작은 물체를 검출하는 이론적 근거를 제시하며, 이의 응용으로 맘모그램에 존재하는 마이크로캘시피케이션을 검출하는 알고리즘을 제안한다. 검출하고자 하는 물체가 가우시안 형태이고 그 스케일이 웨이블릿 변환에 의해 계산된 것과 일치하며, 그 주변의 잡영이 마르코프 프로세스이면, LoG(Laplacian of Gaussian) 웨이블릿은 멀티스케일 매치 필터로 작용하며, 적절한 디테일 이미지를 단순히 이진화함으로써 최적의 검출기를 구현할 수 있다. 그런데, 마이크로캘시피케이션은 정확한 가우시안 형태를 갖지 않고, 게다가 맘모그램의 배경이미지도 마르코프 프로세스라는 가정에서 벗어난다. 이러한 불일치를 해결하기 위하여, 본 논문에서는 멀티스케일 웨이블릿 계수에서 추출한 특징벡터를 Hotelling observer에 입력하여 처리함으로써 이를 보상하고자 하였다.

• Journal of the Korean Institute of Telematics and Electronics B
• /
• v.33B no.1
• /
• pp.144-152
• /
• 1996

In this paper, a pulse-mode digital multilayer neural network with a massively parallel yet compact and flexible network architecture is presented. Algebraicneural operations are replaced by stochastic processes using pseudo-random pulse sequences and simple logic gates are used as basic computing elements. The distributions of the results from the stochastic processes are approximated using the hypergeometric distribution. A statistical model of the noise(error) is developed to estimate the relative accuracy associated with stochastic computing in terms of mean and variance. Numerical character recognition problems are applied to the network to evaluate the network performance and to justify the validity of analytic results based on the developed statistical model. The network architectures are modeled in VHDL using the mixed descriptions of gate-level and register transfer level (RTL). Experiments show that the statistical model successfully predicts the accuracy of the operations performed in the network and that the character classification rate of the network is competitive to that of ordinary Back-Propagation networks.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.16 no.1
• /
• pp.61-67
• /
• 2016

Opportunistic incremental relaying (OIR) system effectively overcomes the degradations caused by the fading of the wireless channel, and efficiently utilizes the wireless resources. Most of the OIR studies, however, assume spatially fixed relays. The user terminals which are usually served as relays move continuously, the assumption that the relays are fixed is not realistic. In this paper, the location of the spatially random user terminals are modeled by the Poisson point process, and the performance of an OIR system is derived. We noticed that the performance of the OIR system improves with the spatially random relays as well as with the fixed relays. Also the intensity of the relays and the transmitting directions toward the destination affect the performances. The performances of the maximal ratio combining (MRC) and the selection combining (SC) at the destination are compared.

• Journal of Korean Ophthalmic Optics Society
• /
• v.16 no.2
• /
• pp.159-165
• /
• 2011

Purpose: In this paper, the reliability and reproductivity of anti-reflection (AR) coating on ophthalmic lens using optical monitoring system (OMS) were investigated. Methods: The random error simulation and RunSheet performance in Essential Macleod software to confirm possibility of AR coating using OMS were performed. The coating process of 19 batches was carried out in order to perform reproductivity test of AR coating after simulation process. Results: As a result, the coating results of 19 batches had shown the excellent reproductivity of about 0.5% error. Conclusions: We confirmed the excellent reproductivity and reliability of AR coating on ophthalmic lens using optical monitoring system from our results.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.24 no.8A
• /
• pp.1187-1194
• /
• 1999

In general wireless mobile channel is modeled as complex random processes having a narrowband spectrum. In this paper, a real-time feneration of fading signals using a DSP chip is described. Real-time simulator is designed so that simulation parameters such as mobile terminal speed, carrier frequency, power ratio of line-of-sight component versus multipath, and variance of received power can be chosen in the window. Design algorithms for the generation of ideal fading signals with a minimum DSP computation and trade-offs are investigated. The accuracy of the statistical characteristics is verified through the comparison of measured results with the theoretical prediction.

• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.38 no.4
• /
• pp.325-334
• /
• 2001

It is shown that the multiscale prewhitening matched filter for detecting Gaussian objects in Markov noise can be implemented by the undecimated wavelet transform with a biorthogonal spline wavelet. If the object to be detected is Gaussian shaped and its scale coincides with one of those computed by the wavelet transform, and if the background noise is truly Markov, then optimum detection is realized by thresholding the appropriate details image. Our detection algorithm is applied to the digitized mammograms for detecting microcalcifications. However, microcalcifications are not exactly Gaussian shaped and its background noise may not be Markov. In order to campensate for these discrepancy, Hotelling observer is employed, which is applied to feature vectors comprised of 3-octave wavelet coefficients.