• Journal of the Institute of Electronics Engineers of Korea TC
• /
• v.37 no.4
• /
• pp.35-43
• /
• 2000

In this paper, we propose to transmit convolutionally coded DS waveforms over orthogonally overlapped subchannels. It is shown that the proposed system, the convolutionally coded orthogonal multicarrier DS/CDMA system, significantly outperforms the system using frequency diversity combining. It is also shown that the proposed system has better performance than the convolutionally coded almost non-overlapped multicarrier DS/CDMA system under the condition that the information rate and total available bandwidth are the same.

• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 2001.05a
• /
• pp.126-130
• /
• 2001

In this paper, we propose the design methodology of target tracking system using fuzzy basis function expansion (FBFE) based on virus evolutionary genetic algorithm(VEGA). In general, the objective of target tracking is to estimate the future trajectory of the target based on the past position of the target obtained from the sensor. In the conventional and mathematical nonlinear filtering method such as extended Kalman filter (EKF), the performance of the system may be deteriorated in highly nonlinear situation. To resolve these problems of nonlinear filtering technique, by appling artificial intelligent technique to the tracking control of moving targets, we combine the advantages of both traditional and intelligent control technique. In the proposed method, after composing training datum from the parameters of extended Kalman filter, by combining FBFE, which has the strong ability for the approximation, with VEGA, which prevent GA from converging prematurely in the case of lack of genetic diversity of population, and by identifying the parameters and rule numbers of fuzzy basis function simultaneously, we can reduce the tracking error of EKF. Finally, the proposed method is applied to three dimensional tracking problem, and the simulation results shows that the tracking performance is improved by the proposed method.

• Animal cells and systems
• /
• v.2 no.2
• /
• pp.281-285
• /
• 1998

We examined Y chromosomal DNA polymorphisms at the DYS19 and DXYS5Y loci in a total of 480 unrelated male samples from the Korean population. All five common alleles were identified at the tetranucleotide microsatellite locus DYS19 in this study. The C allele was the most frequent (212/480), followed by D (136/480), B (75/480), E (36/480) and A (21/480) allele. The frequency of Y2 allele at the DXYS5Y locus was found to be 4.6% (22/480). Combining the allelic variation at these two loci resulted in a total of 9 combination haplotypes. The mean combination haplotype diversity wIns 0.72. Based on the results of these two loci, Korean and Japanese populations may share some common genetic structure that is rare or absent in the other ethnic groups. The genetic similarity between Korean and Japanese populations may be due to the large infusion of Y chromosomes through the Yayoi migration starting 2,300 years ago from Korea to Japan.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.37 no.5C
• /
• pp.444-452
• /
• 2012

The detection performances, in this paper, are derived according to combination of the multiband GNSS signals in broadband jamming environments. The detection probabilities depending on the false alarm probabilities are derived and presented via Monte-Carlo simulation under the assumption as follows: the GNSS signals are perfectly orthogonal and simultaneously received by the receiver using non-coherent correlation.

• Korean Institute of Interior Design Journal
• /
• no.39
• /
• pp.83-91
• /
• 2003

The planning of circulation area and circulation path are very important elements for layout of interior space of housing. This study is, therefore, aimed at synthetical review of that area from a functional, structural and socio-cultural point of view as well as typological analysis of that area. In the interior space of housing the networking of spaces according to the circulation route imply divers aspects of independent or dependant significances. It is thus, closely related with to whether circulation area open or closed, whether circulation path concentrated or distributed, and whether it passes Individual rooms or not. With regard to relationship with public and private spaces, there are many grounds for combining each other. This study also tried to develope program for planning of circulation system, utilizing typological analysis of them. At the same time this study suggests examples for layout of housing spaces. It is expected that the results represented In the form of systematic diagram will deserve to be a tool for providing an appropriate solution to the problem of diversity of user's needs.

• Molecular & Cellular Toxicology
• /
• v.2 no.1
• /
• pp.11-14
• /
• 2006

In this study we attempted to develop a novel genomic method to selectively isolate target functional microbial genomes from environmental samples. For this purpose, stable isotope probing (SIP) was applied in selectively isolating organic pollutant-assimilating populations. When soil microbes were fed with $^{13}C-labeled $ biphenyl, biphenyl-utilizing cells were incorporated with the heavy carbon isotope. The heavy DNA portion was successfully separated by CsCl equilibrium density gradient. And the diversity in the heavy DNA was sufficiently reduced, being suitable for the current DNA microarray techniques to detect biphenyl-utilizing populations in the soil. In addition, we proposed a new way to get more genetic information by combining this SIP method with selective metagenomic approach. The increased selective power of these new DNA isolation methods will be expected to provide a good quality of new genetic information, which, in turn, will result in development of a variety of biomarkers that may be used in assessing ecotoxicology issues including the impacts of organic hazards, and antibiotic-resistant pathogens on human and ecological systems.

• Journal of Korea Multimedia Society
• /
• v.19 no.8
• /
• pp.1636-1645
• /
• 2016

In navigation systems that use collected trajectory for routing, the number and diversity of trajectory data are crucial despite the infeasible limitation which is that all routes should be collected in person. This paper suggests an algorithm combining trajectories only by collected GPS data and generating new routes for solving this problem. Using distance between two trajectories, the algorithm estimates road intersection, in which it also predicts the correlated direction of them with geographical coordinates and makes a decision to combine them by the correlated direction. With combined and generated trajectory data, this combination way allows trajectory-based navigation to guide more and better routes. In our study, this solution has been introduced. However, the ways in which correlated direction is decided and post-process works have been revised to use the sequential pattern of triangles' area GPS information between two trajectories makes in road intersection and intersection among sets comprised of GPS points. This, as a result, reduces unnecessary combinations resulting redundant outputs and enhances the accuracy of estimating correlated direction than before.

• Journal of Biomedical Engineering Research
• /
• v.40 no.2
• /
• pp.48-54
• /
• 2019

This paper proposed a method of validation data augmentation for improving the grading accuracy of diabetic macular edema (DME) using deep learning. The data augmentation technique is basically applied in order to secure diversity of data by transforming one image to several images through random translation, rotation, scaling and reflection in preparation of input data of the deep neural network (DNN). In this paper, we apply this technique in the validation process of the trained DNN, and improve the grading accuracy by combining the classification results of the augmented images. To verify the effectiveness, 1,200 retinal images of Messidor dataset was divided into training and validation data at the ratio 7:3. By applying random augmentation to 359 validation data, $1.61{\pm}0.55%$ accuracy improvement was achieved in the case of six times augmentation (N=6). This simple method has shown that the accuracy can be improved in the N range from 2 to 6 with the correlation coefficient of 0.5667. Therefore, it is expected to help improve the diagnostic accuracy of DME with the grading information provided by the proposed DNN.

• Journal of the Korean Society of Clothing and Textiles
• /
• v.43 no.4
• /
• pp.459-473
• /
• 2019

This study analyzes a new aesthetic value of fashion design utilizing media art technology based on Marshall McLuhan's media theory, vitalize a creative fashion design by applying media art technology into traditional fashion design and discover the possibility of various formative expressions. The result of the analysis under the three criterion extracted through the examination of other art genre such as dance, music, architecture and painting are as follows. First, new concept clothing was designed in the way of combining science and technology with existing costume designs. Second, applying technology based media art images on clothes has been noticeably effected by changes to clothing surfaces. Third, fashion design using technology based media art stimulates the five senses and creates new communication structures. In conclusion, this study reveals that fashion design utilizing technology based media art, an innovative medium for future fashion development in digital society, has expanded the boundaries of fashion design beyond limits and contributed to diversity in creative fashion design.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.14 no.12
• /
• pp.4776-4794
• /
• 2020

We propose a rapid adaptive learning framework for streaming object detection, called EER-ASSL. The method combines the expected error reduction (EER) dependent rollback learning and the active semi-supervised learning (ASSL) for a rapid adaptive CNN detector. Most CNN object detectors are built on the assumption of static data distribution. However, images are often noisy and biased, and the data distribution is imbalanced in a real world environment. The proposed method consists of collaborative sampling and EER-ASSL. The EER-ASSL utilizes the active learning (AL) and rollback based semi-supervised learning (SSL). The AL allows us to select more informative and representative samples measuring uncertainty and diversity. The SSL divides the selected streaming image samples into the bins and each bin repeatedly transfers the discriminative knowledge of the EER and CNN models to the next bin until convergence and incorporation with the EER rollback learning algorithm is achieved. The EER models provide a rapid short-term myopic adaptation and the CNN models an incremental long-term performance improvement. EER-ASSL can overcome noisy and biased labels in varying data distribution. Extensive experiments shows that EER-ASSL obtained 70.9 mAP compared to state-of-the-art technology such as Faster RCNN, SSD300, and YOLOv2.