• Journal of Communications and Networks
• /
• v.7 no.3
• /
• pp.307-312
• /
• 2005

In this work, we study the performance of a serial concatenated scheme comprising a convolutional code (CC) and an orthogonal space-time block code (STBC) separated by an inter-leaver. Specifically, we derive performance bounds for this concatenated scheme, clearly quantify the impact of using a CC in conjunction with a STBC, and compare that to using a STBC code only. Furthermore, we examine the impact of performing antenna selection at the receiver on the diversity order and coding gain of the system. In performing antenna selection, we adopt a selection criterion that is based on maximizing the instantaneous signal-to­noise ratio (SNR) at the receiver. That is, we select a subset of the available receive antennas that maximizes the received SNR. Two channel models are considered in this study: Fast fading and quasi-static fading. For both cases, our analyses show that substantial coding gains can be achieved, which is confirmed through Monte-Carlo simulations. We demonstrate that the spatial diversity is maintained for all cases, whereas the coding gain deteriorates by no more than $10\;log_{10}$ (M / L) dB, all relative to the full complexity multiple-input multiple-output (MIMO) system.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.10 no.2
• /
• pp.574-592
• /
• 2016

Range-free localization algorithm computes a normal node's position by estimating the distance to anchors which know their actual position. In recent years, reliable anchor selection research has been gained a lot of attention because this approach improves localization accuracy by selecting the only subset of anchors called reliable anchor. The distance estimation accuracy and the geometric shape formed by anchors are the two important factors which need to be considered when selecting the reliable anchors. In this paper, we study the relationship between a relative position of three anchors and localization error. From this study, under ideal condition, which is with zero localization error, we find two conditions for anchor selection, thereby proposing a novel anchor selection algorithm that selects three anchors matched most closely to the two conditions, and the validities of the conditions are proved using two theorems. By further employing the conditions, we finally propose a novel range-free localization algorithm. Simulation results show that the proposed algorithm shows considerably improved performance as compared to other existing works.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.18 no.1
• /
• pp.94-99
• /
• 2008

In the pattern classification problem, feature selection is an important technique to improve performance of the classifiers. Particularly, in the case of classifying with a large number of features or variables, the accuracy of the classifier can be improved by using the relevant feature subset to remove the irrelevant, redundant, or noisy data. In this paper we propose a feature selection method using genetic algorithm and information theory. Experimental results show that this method can achieve better performance for pattern recognition problems than conventional ones.

• Journal of Digital Contents Society
• /
• v.17 no.6
• /
• pp.509-521
• /
• 2016

A novel Samples Balanced Genetic Algorithm combined with Extreme Learning Machine (SBGA-ELM) for Parkinson's Disease diagnosis and detecting bio-markers is presented in this paper. Proposed approach uses genes' expression data of 22,283 genes from open source ParkDB data base for accurate PD diagnosis and detecting bio-markers. Proposed SBGA-ELM includes two major steps: feature (genes) selection and classification. Feature selection procedure is based on proposed Samples Balanced Genetic Algorithm designed specifically for genes expression data from ParkDB. Proposed SBGA searches a robust subset of genes among 22,283 genes available in ParkDB for further analysis. In the "classification" step chosen set of genes is used to train an Extreme Learning Machine (ELM) classifier for an accurate PD diagnosis. Discovered robust subset of genes creates ELM classifier with stable generalization performance for PD diagnosis. In this research the robust subset of genes is also used to discover 24 bio-markers probably responsible for Parkinson's Disease. Discovered robust subset of genes was verified by using existing PD diagnosis approaches such as SVM and PBL-McRBFN. Both tested methods caused maximum generalization performance.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2007.05a
• /
• pp.439-442
• /
• 2007

패턴 인식의 성능 향상을 위해 효율적인 특징 선택을 해주는 것은 중요한 과정 중 하나이다. 본 연구에서는 최근에 제시되었던 특징 선택 방법들 중 대표적인 4 가지의 방법을 선택하여 성능을 비교 분석하였다. 데이터의 특징을 줄여주는 기능과 적은 수의 특징으로 더 좋은 패턴 인식 성능을 보여주는 기능의 수행 능력을 중심으로 평가하였으며, 각기 다른 형태의 데이터에 대해 각 방법들이 보이는 성능도 관찰하였다. 이를 토대로 각 방법의 장단점과 특징에 대해 생각해 본다.

• Journal of the Korean Statistical Society
• /
• v.13 no.1
• /
• pp.25-31
• /
• 1984

A technique which has been used for the evaluation of certain kinds of multiple integrals, viz., the technique of imcomplete gamma function operators, is employed and extended to the case where the parameters and arguments are non-equal and non-integer for the probability integral of the inverted Dirichlet distribution. Several types of recurrence formulas have been developed for the tail probabilities and a subset selection procedure in ranking variances is discussed as an application.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2007.11a
• /
• pp.27-29
• /
• 2007

패턴 인식 문제를 푸는데 있어 특징 선택을 해주는 것은 패턴 인식의 성능 향상을 위해 중요한 과정 중 하나이다. 본 연구에서는 대표적인 생태계 모방 알고리즘 2 가지를 선택하여 특징 선택 문제에 적용하여 보고, 그 성능을 비교 분석하였다. 데이터의 특징을 줄여주는 기능과 패턴 인식 성능의 향상 여부를 중심으로 평가하였으며, 이를 통해 생태계 모방 알고리즘이 특징 선택 문제에 효과적으로 사용될 수 있는지에 대해 논의해보고, 두 방법의 장단점과 특징에 대해 생각해 본다.

• Korean Journal of Mathematics
• /
• v.29 no.3
• /
• pp.581-591
• /
• 2021

In this paper, we discuss and investigate the existence of the inclusion L^p(.),𝜃 (𝜇) ⊆ L^q(.),𝜃 (𝜈), where 𝜇 and 𝜈 are two finite measures on (X, Σ). Moreover, we show that the generalized variable exponent grand Lebesgue space L^p(.),𝜃 (Ω) has a potential-type approximate identity, where Ω is a bounded open subset of ℝ^d.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.7 no.4
• /
• pp.82-92
• /
• 1997

In this paper, a fuzzy measure and integral-based model fnr interactive systems is proposed. The processes of model identification consists of the following three steps : (i) structure identification (ii) parameter identification and (iii) selection of an optimal model. An algorithm for the model structure identification using the well-known genetic algorithm ((;A) with a modified selection operator is proposed. A method for the identification of par;imetcrs corresponding to fuzzy measures is presented. A statistical model selection criterion is used for the selection of an optimal model among the candidates. Finally, experimental results obtained hy applying the proposed model to the subjective evaluation data set and the well-known time series data are presented to show the validity of the proposed model.

• Journal of Advanced Marine Engineering and Technology
• /
• v.40 no.5
• /
• pp.437-446
• /
• 2016

Feature selection has become an essential technique to reduce the dimensionality of data sets. Many features are frequently irrelevant or redundant for the classification tasks. The purpose of feature selection is to select relevant features and remove irrelevant and redundant features. Applications of the feature selection range from text processing, face recognition, bioinformatics, speaker verification, and medical diagnosis to financial domains. In this study, we focus on filter methods based on information entropy : IG (Information Gain), FCBF (Fast Correlation Based Filter), and mRMR (minimum Redundancy Maximum Relevance). FCBF has the advantage of reducing computational burden by eliminating the redundant features that satisfy the condition of approximate Markov blanket. However, FCBF considers only the relevance between the feature and the class in order to select the best features, thus failing to take into consideration the interaction between features. In this paper, we propose an improved FCBF to overcome this shortcoming. We also perform a comparative study to evaluate the performance of the proposed method.