• The Journal of the Acoustical Society of Korea
• /
• v.22 no.7
• /
• pp.603-611
• /
• 2003

This paper proposes an effective feature compensation scheme based on speech model for achieving robust speech recognition. Conventional model-based method requires off-line training with noisy speech database and is not suitable for online adaptation. In the proposed scheme, we can relax the off-line training with noisy speech database by employing the parallel model combination technique for estimation of correction factors. Applying the model combination process over to the mixture model alone as opposed to entire HMM makes the online model combination possible. Exploiting the availability of noise model from off-line sources, we accomplish the online adaptation via MAP (Maximum A Posteriori) estimation. In addition, the online channel estimation procedure is induced within the proposed framework. For more efficient implementation, we propose a selective model combination which leads to reduction or the computational complexities. The representative experimental results indicate that the suggested algorithm is effective in realizing robust speech recognition under the combined adverse conditions of additive background noise and channel distortion.

• Transactions of the Korean Society of Mechanical Engineers B
• /
• v.24 no.5
• /
• pp.652-659
• /
• 2000

The purpose of this study is to present the most effective numerical scheme to calculate the unsteady flows. In order to calculate the flow quantities of flow past a circular cylinder, Three-time level and five convective schemes are applied to unsteady and convective terms, respectively. The values obtained are compared with those from the existing experimental and numerical studies. At Reynolds numbers up to 160, time intervals can be expanded 10 times of Implicit Euler scheme using Three-time level method, and it is found that QUICK and CUI schemes work much stable than others even if less grid density conditions. The combination of Three-time level and QUICK scheme gives high resolutions for laminar unsteady problems with PC level.

• Journal of applied mathematics & informatics
• /
• v.36 no.1_2
• /
• pp.135-154
• /
• 2018

In this paper, an almost second order overlapping Schwarz method for singularly perturbed third order convection-diffusion type problem is constructed. The method splits the original domain into two overlapping subdomains. A hybrid difference scheme is proposed in which on the boundary layer region we use the combination of classical finite difference scheme and central finite difference scheme on a uniform mesh while on the non-layer region we use the midpoint difference scheme on a uniform mesh. It is shown that the numerical approximations which converge in the maximum norm to the exact solution. We proved that, when appropriate subdomains are used, the method produces convergence of second order. Furthermore, it is shown that, two iterations are sufficient to achieve the expected accuracy. Numerical examples are presented to support the theoretical results. The main advantages of this method used with the proposed scheme are it reduce iteration counts very much and easily identifies in which iteration the Schwarz iterate terminates.

• Journal of Internet Computing and Services
• /
• v.12 no.2
• /
• pp.1-8
• /
• 2011

In this paper, we propose an effective method for cooperative spectrum sensing in cognitive radios where cognitive user(CR) with the highest reliability sensing data is only selected and allowed to report its local decision to FC as only decision making node. The proposed scheme enables CR users to implicitly compare their sensing data reliabilities based on their likelihood ratio, without any collaboration among cognitive radio users. Due to the mechanism, the proposed cooperative scheme can achieves a high spectrum sensing performance while only requiring extremely low cooperation resources such as signaling overhead and cooperative time in comparison with other existing methods such as maximum ratio combination (MRC) based, equal gain combination (EGC) based and conventional hard combination based cooperative sensing methods.

• Nuclear Engineering and Technology
• /
• v.54 no.4
• /
• pp.1526-1533
• /
• 2022

Small modular reactors have garnered considerable attention in the recent years. Plate fuel elements exhibit a good application prospect in small modular pressurized water reactors for marine applications. Further, improved economic benefits can be achieved by extending the core lifetime of small modular reactors. However, it is necessary to realize a large initial residual reactivity for achieving a relatively long burnup depth finally. Thus, the selection of a suitable burnable poison (BP) is a crucial factor that should be considered in the design of small modular reactors. In this study, some candidate BPs are selected to realize the effective control of reactivity. The results show that ²³¹Pa₂O₃, ²⁴⁰Pu₂O₃, ¹⁶⁷Er₂O₃, PACS-J, and PACS-L are ideal candidates of BP, and since the characteristics of BP can increase the final burnup depth of assembly, the economic benefits are gained. Additionally, an optimal combination scheme of BPs is established. Specifically, it is proved that through a reasonable combination of BPs, a low reactivity fluctuation during the lifetime can be achieved, leading to a large final burnup depth.

• ETRI Journal
• /
• v.35 no.3
• /
• pp.491-500
• /
• 2013

A connected component analysis from a binary image is a popular character segmentation method but occasionally fails to segment the characters owing to image noise and uneven illumination. A multimethod binarization scheme that incorporates two or more binary images is a novel solution, but selection of binarization methods has never been analyzed before. This paper reveals the best combination of binarization methods and parameters and presents an in-depth analysis of the multimethod binarization scheme for better character segmentation. We carry out an extensive quantitative evaluation, which shows a significant improvement over conventional single-method binarization methods. Experiment results of six binarization methods and their combinations with different test images are presented.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2020.11a
• /
• pp.144-146
• /
• 2020

Discrete Wavelet Transform (DWT) is an effective technique that is commonly used for detecting noise in collected data of an individual sensor. In addition, the detection accuracy can be significant improved by exploiting the correlation in the data of neighboring sensors of Wireless Sensor Networks (WSNs). Principal component analysis is the powerful technique to analyze the correlation in the multivariate data. In this paper, we propose a DWT-PCA combination scheme for noise detection (DWT-PCA-ND). Experimental results on a real dataset show a remarkably higher performance of DWT-PCA-ND comparing to conventional PCA scheme in detection of noise that is a popular anomaly in collected data of WSN.

• The Transactions of the Korea Information Processing Society
• /
• v.3 no.4
• /
• pp.908-914
• /
• 1996

A number of techniques have been studied for handling uncertainty in the development of expert systems. One of techniques adopted in many expert systems is the Dumpster-Shafer Evidence combination scheme. This has been the main focus among others due to is favorable features and computational complexity. In this paper, we develop and algorithm to deal with the exponential complexity inherent in Dempster-Shafer evidence combination. In the evidence combination process, we divide the frame of discernment into two groups, one for those common in both belief functions and the other for the rest. A property is found that in computing new belief function for the latter group, the result of evidence combination show linear change. The irrelevancy factor is derived and used to compute the change. The main idea of the method is to reduce the size of the frame of discernment and thus exponential complexity.

• IEIE Transactions on Smart Processing and Computing
• /
• v.2 no.1
• /
• pp.1-10
• /
• 2013

In JPEG2000, the Cohen-Daubechies-Feauveau (CDF) 9/7-tap wavelet filter is implemented using the conventional lifting scheme. On the other hand, this wavelet filter has two problems: the filter coefficients remain complex, and the conventional lifting scheme does not consider the image edges in the coding process. This paper proposes an effective lifting scheme to solve these problems. For this purpose, optimal 9/7-tap wavelet filters were designed in two steps. In the first step, the appropriate filter coefficients were selected. In the second step, a median operator was employed to consider the image edges. The experimental results with the median lifting scheme and the combination of filter optimization with the median lifting show that the proposed methods outperform the well-known CDF 9/7-tap wavelet filter of JPEG2000 on the edge-dominant images.

• Journal of the Korean Data and Information Science Society
• /
• v.25 no.2
• /
• pp.431-438
• /
• 2014

Type I hybrid censoring scheme is the combination of the Type I and Type II censoring scheme introduced by Epstein (1954). Epstein considered a hybrid censoring sampling scheme in which the life testing experiment is terminated at a random time $T^*$ which is the time that happens rst among the following two; time of the kth unit is observed or time of the experiment length set in advance. The likelihood function of this scheme from the Rayleigh distribution cannot be solved in a explicit solution and thus we approximate the function by the Taylor series expansion. In this process, we propose four dierent methods of expansion skill.