• Journal of Information Processing Systems
• /
• v.20 no.4
• /
• pp.458-476
• /
• 2024

To address the issue of class imbalance in network traffic data, which affects the network intrusion detection performance, a combined framework using transformers is proposed. First, Tomek Links, SMOTE, and WGAN are used to preprocess the data to solve the class-imbalance problem. Second, the transformer is used to encode traffic data to extract the correlation between network traffic. Finally, a hybrid deep learning network model combining a bidirectional gated current unit and deep neural network is proposed, which is used to extract long-dependence features. A DNN is used to extract deep level features, and softmax is used to complete classification. Experiments were conducted on the NSLKDD, UNSWNB15, and CICIDS2017 datasets, and the detection accuracy rates of the proposed model were 99.72%, 84.86%, and 99.89% on three datasets, respectively. Compared with other relatively new deep-learning network models, it effectively improved the intrusion detection performance, thereby improving the communication security of network data.

• Journal of the Korean Society of Industry Convergence
• /
• v.4 no.1
• /
• pp.35-40
• /
• 2001

A spectral measurement system for reflection and transmission properties by using an optical fiber and an ellipsoidal mirror was newly developed. The hemispherical reflectance and transmittance spectra of several heating resisting ceramics materials were measured from visible to middle infrared region. The directional characteristics of reflection and transmission were also investigated in consideration of the absorptance. The measured data were analyzed by using a four flux model of radiation transfer, The radiation properties could be estimated by the obtained scattering and absorption coefficient spectra.

• Journal of Advanced Marine Engineering and Technology
• /
• v.27 no.4
• /
• pp.535-541
• /
• 2003

This paper describes numerical analysis of the impulse turbine with fixed guide vanes, a high performance hi-directional air turbine having simple structure for wane energy conversion. A 3-dimensional incompressible viscous flow numerical analysis based on the full Reynolds-averaged Wavier-Stokes equations was made to investigate the internal flow behavior Numerical results ate compared with experimental data. As a result, a suitable choice for the one of design factors has been clarified.

• 제어로봇시스템학회:학술대회논문집
• /
• 2002.10a
• /
• pp.34.6-34
• /
• 2002

1. Introduction 2. System Configuration 2.1 Control System 2.1.1 Longitudinal control 2.1.2 Lateral control 2.2 Sensor System 2.2.1 Photo interrupt 2.2.2 Ultrasonic sensor 2.3 Vision system 2.4 Communication system 2.4.1 Data communication 2.4.2 Image Communication 3. Test and Result 3.1 Vision test 3.2 Ultrasonic sensor test 4. Conculsion. Acknowledgment References.

• 한국전산유체공학회:학술대회논문집
• /
• 2003.10a
• /
• pp.36-37
• /
• 2003

Three-dimensional numerical analysis of the flow around rectangular cylinders with various side ratios, D/H, from 0.2 to 2.0 is carried out for Reynolds number of 1000 by using multi-directional finite difference method in multi-grid. The predicted results are well compared with the experimental data. It is found that fluid dynamics characteristics alternate between high pressure mode. and low pressure mode of the base pressure for rectangular cylinder of D/H=0.2-0.6.

• Proceedings of the IEEK Conference
• /
• 2003.07a
• /
• pp.133-136
• /
• 2003

In this paper, we propose a multicast routing protocol for the seamless delivery of multicast data to mobile hosts through the optimal route in IP based mobile networks. The proposed multicast routing protocol is a hybrid method employing the merits of the bi-directional tunneling and the remote subscription by considering the mobility of mobile hosts. The proposed protocol satisfies the bound of end-to-end delay and supports the seamless handoff. The simulation results show that the proposed protocol has better performance in the number of multicast tree reconstruction and tunneling length and packet loss time than the previous protocols.

• KIPS Transactions on Software and Data Engineering
• /
• v.3 no.10
• /
• pp.397-406
• /
• 2014

In order to answer the questions successfully on behalf of the human in DeepQA environments such as Jeopardy! of the American quiz show, the computer is required to have the capability of fast temporal and spatial reasoning on a large-scale commonsense knowledge base. In this paper, we present a scalable spatial reasoning algorithm for deriving efficiently new directional and topological relations using the MapReduce framework, one of well-known parallel distributed computing environments. The proposed reasoning algorithm assumes as input a large-scale spatial knowledge base including CSD-9 directional relations and RCC-8 topological relations. To infer new directional and topological relations from the given spatial knowledge base, it performs the cross-consistency checks as well as the path-consistency checks on the knowledge base. To maximize the parallelism of reasoning computations according to the principle of the MapReduce framework, we design the algorithm to partition effectively the large knowledge base into smaller ones and distribute them over multiple computing nodes at the map phase. And then, at the reduce phase, the algorithm infers the new knowledge from distributed spatial knowledge bases. Through experiments performed on the sample knowledge base with the MapReduce-based implementation of our algorithm, we proved the high performance of our large-scale spatial reasoner.

• The Journal of the Acoustical Society of Korea
• /
• v.24 no.8
• /
• pp.469-480
• /
• 2005

There have been always some difficulties in target setting and conditioning of acoustic performances or the Korean traditional music hall due mainly to the lack of the information on the sound radiation characteristics of Korean musical sources. As the 2nd experiment succeeding the previous study[1], the radiation characteristics of eight typical Korean traditional musical sources were investigated if precision. The selected musical sources were Geomungo, Haegeum (string), Piri, Taepyeongso (woodwind), Buk, Kwaengguari, Jing (drum), and male Pansori Chang (vocal Performance). The results show that the directivity pattern of each instrument is different and has their own directivity characteristics. Measured directional and spectral characteristics of traditional Korean music sources were implemented into the computation of architectural acoustic measures. Significant differences in the acoustic measures at receiver positions were observed between the results in using the omni-directional source and the directional one. In order to investigate the acoustical characteristics of the instruments depending on the spatial variation four different shapes of halls were introduced including rectangular, fan. horse-shoe and geometrical shapes. Room acoustical parameters such as RT, SPL, C80, LF, STI were calculated at each type or hall. As the results, It was found that the rectangular hall has the most high clarity. lateral energy and STI values among low shapes of halls. It is thought that the suggested source data and design method can be used as a basic reference in the future acoustic design of performance halls for the Korean traditional music.

• Tunnel and Underground Space
• /
• v.33 no.3
• /
• pp.150-168
• /
• 2023

The anisotropic behavior of rocks is primarily attributed to the directional arrangement of rock-forming minerals and the distribution characteristics of microcracks. Notably, sedimentary and metamorphic rocks often exhibit distinct transverse isotropy in terms of their strength and deformation characteristics. Consequently, it is crucial to gain accurate insights into the deformation and failure characteristics of transversely isotropic rocks during rock mechanics design processes. The deformation of such rocks is described by five independent elastic constants, which are determined through laboratory testing. In this study, the characteristics of the directional variation of apparent elastic constants in transversely isotropic rocks were investigated using experimental data reported in the literature. To achieve this, the constitutive equation proposed by Mehrabadi & Cowin was introduced to calculate the apparent elastic constants more efficiently and systematically in a rotated Cartesian coordinate system. Four transversely isotropic rock types from the literature were selected, and the influence of changes in the orientation of the weak plane on the variations of the apparent elastic modulus, apparent shear modulus, and apparent Poisson's ratio was analyzed. Based on the investigation, a new constraint on the elastic constants has been proposed. If the proposed constraint is satisfied, the directional variation of the apparent elastic constants in transversely isotropic rocks aligns with intuitive predictions of their tendencies.

• International Journal of Aeronautical and Space Sciences
• /
• v.17 no.1
• /
• pp.64-72
• /
• 2016

The coverage area of a GNSS regional ionospheric delay model is mainly determined by the distribution of GNSS ground monitoring stations. Extrapolation of the ionospheric model data can extend the coverage area. An extrapolation algorithm, which combines observed ionospheric delay with the environmental parameters, is proposed. Neural network and least square regression algorithms are developed to utilize the combined input data. The bi-harmonic spline method is also tested for comparison. The IGS ionosphere map data is used to simulate the delays and to compute the extrapolation error statistics. The neural network method outperforms the other methods and demonstrates a high extrapolation accuracy. In order to determine the directional characteristics, the estimation error is classified into four direction components. The South extrapolation area yields the largest estimation error followed by North area, which yields the second-largest error.