• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 2003.09a
• /
• pp.595-598
• /
• 2003

A new classifier is constructed by using a generalized regression neural network (GRNN) in conjunction with a random generator (RC). The RG played a role of generating a number of sets of random spreads given a range for gaussian functions in the pattern layer, The range experimentally varied from 0.4 to 1.4. The DNA sequences consisted 4 types of promoters. The performance of classifier is examined in terms of total classification sensitivity (TCS), and individual classification sensitivity (ICS). for comparisons, another GRNN classifier was constructed and optimized in conventional way. Compared GRNN, the RG-GRNN demonstrated much improved TCS along with better ICS on average.

• IE interfaces
• /
• v.12 no.3
• /
• pp.401-405
• /
• 1999

We present a simulation model to generate the spatial distribution of mobiles in cellular communication network. Three types of spatial distributions are considered; biased, random, and ratio-based distributions. This study also points out and corrects the critical errors performed by Das and Morgera(1997) in getting random location of mobiles. By applying a simple path loss model, the effects of our correction on the signal-to-interference(SIR) ratio are discussed. The numerical results indicate that the variation of SIR in the Das's biased distribution is larger than that of other distributions. As compared with the random distribution, the average SIR error of the biased distribution is 91.1%.

• Journal of the Korea Society of Computer and Information
• /
• v.22 no.3
• /
• pp.61-67
• /
• 2017

This paper analyzes the robustness of the network based on the centrality of vertices in the graph. In this paper, a random graph is generated and a modified graph is constructed by adding or removing vertices or edges in the generated random graph. And then we analyze the robustness of the graph by observing changes in the centrality of the random graph and the modified graph. In the process modifying a graph, we changes some parts of the graph, which has high values of centralities, not in the whole. We study how these additional changes affect the robustness of the graph when changes occurring a group that has higher centralities than in the whole.

• Journal of Information Processing Systems
• /
• v.7 no.1
• /
• pp.17-28
• /
• 2011

Despite the fact that the deployment of sensor networks and target tracking could both be managed by taking full advantage of Voronoi diagrams, very little few have been made in this regard. In this paper, we designed an optimized barrier coverage and an energy-efficient clustering algorithm for forming Vonoroi-based Wireless Sensor Networks(WSN) in which we proposed a mobile target tracking scheme (CTT&MAV) that takes full advantage of Voronoi-diagram boundary to improve detectability. Simulations verified that CTT&MAV outperforms random walk, random waypoint, random direction and Gauss-Markov in terms of both the average hop distance that the mobile target moved before being detected and lower sensor death rate. Moreover, we demonstrate that our results are robust as realistic sensing models and also validate our observations through extensive simulations.

• ETRI Journal
• /
• v.28 no.3
• /
• pp.311-319
• /
• 2006

This paper presents a terminal-assisted frame-based packet reservation multiple access (TAF-PRMA) protocol, which optimizes random access control between heterogeneous traffic aiming at more efficient voice/data integrated services in dynamic reservation TDMA-based broadband access networks. In order to achieve a differentiated quality-of-service (QoS) guarantee for individual service plus maximal system resource utilization, TAF-PRMA independently controls the random access parameters such as the lengths of the access regions dedicated to respective service traffic and the corresponding permission probabilities, on a frame-by-frame basis. In addition, we have adopted a terminal-assisted random access mechanism where the voice terminal readjusts a global permission probability from the central controller in order to handle the 'fair access' issue resulting from distributed queuing problems inherent in the access network. Our extensive simulation results indicate that TAF-PRMA achieves significant improvements in terms of voice capacity, delay, and fairness over most of the existing medium access control (MAC) schemes for integrated services.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.41 no.1
• /
• pp.104-106
• /
• 2016

In this letter, we consider a random access scheme for underwater acoustic network, in which a slotted ALOHA with variable slot length is designed to enhance the random access performance for the nodes with the varying propagation delay.

• Korea Trade Review
• /
• v.46 no.2
• /
• pp.281-299
• /
• 2021

This paper analyzes the feasibility of using machine learning methods to forecast the employment. The machine learning methods, such as decision tree, artificial neural network, and ensemble models such as random forest and gradient boosting regression tree were used to forecast the employment in Busan regional economy. The following were the main findings of the comparison of their predictive abilities. First, the forecasting power of machine learning methods can predict the employment well. Second, the forecasting values for the employment by decision tree models appeared somewhat differently according to the depth of decision trees. Third, the predictive power of artificial neural network model, however, does not show the high predictive power. Fourth, the ensemble models such as random forest and gradient boosting regression tree model show the higher predictive power. Thus, since the machine learning method can accurately predict the employment, we need to improve the accuracy of forecasting employment with the use of machine learning methods.

• Journal of Internet Computing and Services
• /
• v.12 no.6
• /
• pp.63-71
• /
• 2011

Cardiac disorder data are generally tested using the classifier and QRS-Complex and R-R interval which is used in this experiment are often extracted by ECG(Electrocardiogram) signals. The experimentation of ECG data with classifier is generally performed with SVM(Support Vector Machine) and MLP(Multilayer Perceptron) classifier, but this study experimented with Best-First Decision Tree(B-F Tree) derived from the Dicision Tree among Random Forest classifier algorithms to improve accuracy. To compare and analyze accuracy, experimentation of SVM, MLP, RBF(Radial Basic Function) Network and Decision Tree classifiers are performed and also compared the result of announced papers carried out under same interval and data. Comparing the accuracy of Random Forest classifier with above four ones, Random Forest is the best in accuracy. As though R-R interval was extracted using Band-pass filter in pre-processing of this experiment, in future, more filter study is needed to extract accurate interval.

• The KIPS Transactions:PartC
• /
• v.10C no.7
• /
• pp.899-914
• /
• 2003

It is generally accepted that self-similar (or fractal) processes may provide better models for teletraffic in modern telecommunication networks than Poisson Processes. If this is not taken into account, it can lead to inaccurate conclusions about performance of telecommunication networks. Thus, an important requirement for conducting simulation studies of telecommunication networks is the ability to generate long synthetic stochastic self-similar sequences. Three generators of pseudo-random self-similar sequences, based on the FFT〔20〕, RMD〔12〕 and SRA methods〔5, 10〕, are compared and analysed in this paper. Properties of these generators were experimentally studied in the sense of their statistical accuracy and times required to produce sequences of a given (long) length. While all three generators show similar levels of accuracy of the output data (in the sense of relative accuracy of the Horst parameter), the RMD- and SRA-based generators appear to be much faster than the generator based on FFT. Our results also show that a robust method for comparative studies of self-similarity in pseudo-random sequences is needed.

• Journal of Communications and Networks
• /
• v.3 no.4
• /
• pp.334-341
• /
• 2001

The provision of QoS control is a key of the successful deployment of multimedia networks. Buffer management plays an important role in QoS control. Therefore, this paper proposes a novel QoS buffer management algorithm named GREEN (Global Random Early Estimation for Nipping), which extends the concepts of ERD (early random drop) and RED (random early detection). Specifically, GREEN enhances the concept of "Random" to "Global Random" by globally considering the random probability function. It also enhances the concept of "Early" to "Early Esti mation" by early estimating the network status. For performance evaluation, except compared with RED, extensive simulation cases are performed to probe the characteristics of GREEN.