• 한국통신학회논문지
• /
• 제37A권10호
• /
• pp.809-814
• /
• 2012

본 논문은 멀티레벨을 가지는 홀로그래픽 저장 장치에서 EM (Expectation-maximization) 알고리즘을 이용한 적응 문턱전압검출기를 제안한다. 멀티레벨을 이용한 홀로그래픽 저장 장치의 경우 DC 오프셋의 정도에 따라 비적응 문턱전압검출기의 성능에 매우 심각한 영향을 미친다. EM 방법은 채널을 통과한 데이터를 이용해 Expectation step과 maximization step을 반복하면서 평균과 분산을 추정하는 방법이다. DC 오프셋이 있는 상황에서 제안된 방법을 적용하여 문턱값을 찾아내서 검출한 결과 일정한 한도 내의 DC 오프셋의 경우는 DC 오프셋이 없는 경우와 동일한 성능을 보였다.

• 한국경영과학회지
• /
• 제30권2호
• /
• pp.63-71
• /
• 2005

In the field of nuclear reactor safety study, common cause failures (CCFs) became significant contributors to system failure probability and core damage frequency in most Probabilistic risk assessments. However, it is hard to estimate the reliability of such a system, because of the dependency of components caused by CCFs. In order to analyze the system, we propose an analytic method that can find the parameters with lack of raw data. This study adopts the shock model in which the failure probability increases as the shock is cumulated. We use two-step Expectation and Maximization (EM) algorithm to find the unknown parameters. In order to verify the analysis result, we perform the simulation under same environment. This approach might be helpful to build the defensive strategy for the CCFs.

• Journal of Communications and Networks
• /
• 제12권4호
• /
• pp.317-329
• /
• 2010

In recent years, many sparse estimation methods, also known as compressed sensing, have been developed. However, most of these methods presume that the measurement matrix is completely known. We develop a new blind maximum likelihood method-the expectation-sparse-maximization (ESpaM) algorithm-for models where the measurement matrix is the product of one unknown and one known matrix. This method is a variant of the expectation-maximization algorithm to deal with the resulting problem that the maximization step is no longer unique. The ESpaM algorithm is justified theoretically. We present as well numerical results for two concrete examples of blind channel identification in digital communications, a doubly-selective channel model and linear time invariant sparse channel model.

• IEIE Transactions on Smart Processing and Computing
• /
• 제4권4호
• /
• pp.202-208
• /
• 2015

In this study, we propose a new inference algorithm for a multiclass Gaussian process classification model using a variational EM framework and the Laplace approximation (LA) technique. This is performed in two steps, called expectation and maximization. First, in the expectation step (E-step), using Bayes' theorem and the LA technique, we derive the approximate posterior distribution of the latent function, indicating the possibility that each observation belongs to a certain class in the Gaussian process classification model. In the maximization step, we compute the maximum likelihood estimators for hyper-parameters of a covariance matrix necessary to define the prior distribution of the latent function by using the posterior distribution derived in the E-step. These steps iteratively repeat until a convergence condition is satisfied. Moreover, we conducted the experiments by using synthetic data and Iris data in order to verify the performance of the proposed algorithm. Experimental results reveal that the proposed algorithm shows good performance on these datasets.

• Journal of Advanced Marine Engineering and Technology
• /
• 제21권3호
• /
• pp.293-301
• /
• 1997

Edge detection is the first step and very important step in image analysis. We cast edge detec¬tion as a problem in cost maximization. This is acheived by the formulation of a cost function that evaluates the quality of edge configurations. The cost function can be used as a basis for compar¬ing the performances of different detectors. We used a Genetic Algorithm for maximizing cost func¬tion. Genetic algorithms are a class of adaptive search techniques that have been intensively stud¬ied in recent years and have been prone to converge prematurely before the best solution has been found. This paper shows that carefully chosen modifications(three factors of the crossover opera¬tor) are implemented can be effective in alleviating this problem.

• 한국통신학회논문지
• /
• 제34권4C호
• /
• pp.397-402
• /
• 2009

대용량의 데이터 전송을 위해 최근에 많이 사용되고 있는 OFDM 시스템은 다중경로페이딩에는 유리하지만 시간 선택성 페이딩에서는 자유롭지 못하고 이를 위해 채널추정이 필요하게 된다. 이를 위해 본 논문에서는 기존의 EM 알고리즘의 성능을 향상시키기 위해 LPC 알고리즘을 접목시킨 EM-LPC 알고리즘을 제안했다. 이 알고리즘은 기존의 EM 알고리즘의 M-Step 다음에 LPC 알고리즘을 사용하여 미리 다음 pilot 데이터를 추정할 수 있게된다. 또한 interpolation을 통한 데이터 수정도 가능하게 된다. 살펴본 결과 EM-LPC 알고리즘이 기존의 LS 알고리즘과 EM 알고리즘보다 좋은 성능을 내는 것을 볼 수 있었다. 또한 EM 루프의 반복도 역시 줄어드는 것을 볼 수 있었다.

• Journal of Electrical Engineering and Technology
• /
• 제11권3호
• /
• pp.670-674
• /
• 2016

The control performance of hybrid PWM inverter using a phase current measurement is presented in this paper. The hybrid PWM technique consists of space vector pulse width modulation (SVPWM) and six-step voltage control operation. The SVPWM is performed to reduce the harmonic components in the low speed region, and the six-step modulation is applied to increase the maximum speed of the IPMSM in the high speed region. Therefore, it is possible to obtain a great performance in both the low speed range and high speed range. However, the six-step modulation cannot be completely implemented, since the inverter that includes the lag-shunt sensing method has an immeasurable current region. In this paper, a quasi-six-step modulation using a modified voltage vector is proposed. The validity and usefulness of the proposed PWM technique is verified by MATLAB/Simulink and experimental results.

• 한국군사과학기술학회지
• /
• 제12권1호
• /
• pp.97-105
• /
• 2009

In this paper, the performance of AP(Alternating Projection) and EM(Expectation Maximization) algorithms is investigated in terms of detection of multiple signals, resolvability of coherent signals and the efficiency of sensor array processing. The basic idea of these algorithms is utilization of relaxation technique of successive 1D maximization to solve a direction finding problem by maximizing the multidimensional likelihood function. It means that the function is maximized over only for a single parameter while the other parameters are fixed at each step of the iteration. According to simulation results, the algorithms showed good performance for both incoherent and coherent multiple signals. Moreover, some advantages are identified for direction finding with very small samples and fast convergence. The performance of AP algorithm is compared with that of EM using multiple criteria such as the number of sensor, SNR, the number of samples, and convergence speed over uniform circular array. It is resulted AP algorithm is superior to EM overally except for one criterion, convergence speed. Especially, for EM algorithm there is no performance difference between incoherent and coherent case. In conclusion, AP and EM are viable and practical alternatives, which can be applied to a direction under due to the resolvability of multi-path signals, reliable performance and no troublesome eigen-decomposition of the sample-covariance matrix.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제13권5호
• /
• pp.2299-2318
• /
• 2019

Location-aware networks are of great importance for both civil lives and military applications. Methods based on line-of-sight (LOS) measurements suffer sever performance loss in harsh environments such as indoor scenarios, where sensors can receive both LOS and non-line-of-sight (NLOS) measurements. In this paper, we propose a data association (DA) process based on the expectation maximization (EM) algorithm, which enables us to exploit multipath components (MPCs). By setting the mapping relationship between the measurements and scatters as a latent variable, coefficients of the Gaussian mixture model are estimated. Moreover, considering the misalignment of sensor position, we propose a space-alternating generalized expectation maximization (SAGE)-based algorithms to jointly update the target localization and sensor position information. A two dimensional (2-D) circularly symmetric Gaussian distribution is employed to approximate the probability density function of the sensor's position uncertainty via the minimization of the Kullback-Leibler divergence (KLD), which enables us to calculate the expectation step with low computational complexity. Moreover, a distributed implementation is derived based on the average consensus method to improve the scalability of the proposed algorithm. Simulation results demonstrate that the proposed centralized and distributed algorithms can perform close to the Monte Carlo-based method with much lower communication overhead and computational complexity.

• 산업공학
• /
• 제14권4호
• /
• pp.356-364
• /
• 2001

Production Improvement Method in TOC consists of five steps, but it is very difficult for most firms to implement it because it lacks the detailed methods at each step. This paper suggests some of detailed methods to implement the TOC. In the first step, computer simulation is used to identify the constraints in production lines. Subsequently, ASP, AUT, and CM calculation are defined for the second step, which are helpful to exploit the company's constraints. We also suggest the OEE method to effectively exploit the constraints of production lines in the factory. Finally the TOC/OEE procedure is suggested to optimize the investment in the fourth step. As an illustrative example, we introduce a case of a wafer manufacturer to adopt the suggested methods. The benefits of implementating the suggested methods are addressed in the framework of the balanced scorecard.