• Annual Conference of KIPS
• /
• 2021.11a
• /
• pp.61-63
• /
• 2021

Machine learning model gained the popularity in recent years as multi-national companies have incorporated machine learning in their services. Such service is called machine learning as a service (MLaSS). Such services are provided to users based on charge-per-query which triggers the motivations for adversaries to steal the trained victim model to reduce the cost of using the service. Therefore, it is important for companies that provide MLaSS to protect their intellectual property (IP) against adversaries. It has been arms race between the attack and defence in a context of the privacy of machine learning models. In this paper, we provide a comprehensive study of recent development in protecting privacy of machine learning models.

• International Journal of Computer Science & Network Security
• /
• v.23 no.11
• /
• pp.190-194
• /
• 2023

By looking the importance of communication, data delivery and access in various sectors including governmental, business and individual for any kind of data, it becomes mandatory to identify faults and flaws during cyber communication. To protect personal, governmental and business data from being misused from numerous advanced attacks, there is the need of cyber security. The information security provides massive protection to both the host machine as well as network. The learning methods are used for analyzing as well as preventing various attacks. Machine learning is one of the branch of Artificial Intelligence that plays a potential learning techniques to detect the cyber-attacks. In the proposed methodology, the Decision Tree (DT) which is also a kind of supervised learning model, is combined with the different cross-validation method to determine the accuracy and the execution time to identify the cyber-attacks from a very recent dataset of different network attack activities of network traffic in the UNSW-NB15 dataset. It is a hybrid method in which different types of attributes including Gini Index and Entropy of DT model has been implemented separately to identify the most accurate procedure to detect intrusion with respect to the execution time. The different DT methodologies including DT using Gini Index, DT using train-split method and DT using information entropy along with their respective subdivision such as using K-Fold validation, using Stratified K-Fold validation are implemented.

• Journal of the Korean Society of Visualization
• /
• v.22 no.1
• /
• pp.49-60
• /
• 2024

The purpose of this study establishes the deep neural network (DNN) and Decoder models to predict the flow and thermal fields of three-dimensional wavy wings as a passive flow control. The wide ranges of the wavy geometric parameters of wave amplitude and wave number are considered for the various the angles of attack and the aspect ratios of a wing. The huge dataset for training and test of the deep learning models are generated using computational fluid dynamics (CFD). The DNN and Decoder models exhibit quantitatively accurate predictions for aerodynamic coefficients and Nusselt numbers, also qualitative pressure, limiting streamlines, and Nusselt number distributions on the surface. Particularly, Decoder model regenerates the important flow features of tiny vortices in the valleys, which makes a delay of the stall. Also, the spiral vortical formation is realized by the Decoder model, which enhances the lift.

• Proceedings of the KSME Conference
• /
• 2007.05b
• /
• pp.2857-2861
• /
• 2007

Flow and aerodynamic characteristics were analyzed numerically for a commercial passenger airplane, Boeing 747-400, flying in the cruising condition. The model geometry with 100:1 in scale was obtained by the photo scanning measurement with the maximum error of 1.4% comparing with the real airplane dimension. The three-dimensional inviscid steady compressible governing equations were solved by the finite volume method in the unstructured grid system. The convective terms were treated by the Crank-Nicholson and first-order upwind schemes. In the computational results, the strong wing-tip vortices were clearly observed and the pressure contours on the airplane surface were suggested. The lift and drag forces in the wing with engines increase by 1.49% and 3.9%, respectively compared with the case without engines. The aerodynamic forces were estimated quantitatively for each element which consists of the airplane.

• Journal of National Security and Military Science
• /
• s.1
• /
• pp.391-421
• /
• 2003

While the social activities using Internet become generalized, the side effect of the information security violation is increasing steadily and threaten the countries which is not ready to prevent from offensive penetration such as the Information-fighter or Cyber-military. In this paper, we define the concept and characteristics of the modern Information-Warfare and analyze various kinds of threatened elements and also examine the recent trend in other countries. And introducing Computer Forensics raised recently for the confrontation against the security violation in the future, we will show the developing strategies and the necessity in order to response cyber attacks. These developing strategies can be used to ensure and re-trace the technical evidence for the security violation and to achieve the disaster relief effectively. So we hope that can apply them to the actual preparation through developing cyber trial test of the defense and attack for the Information-Warfare.

• International Journal of Aeronautical and Space Sciences
• /
• v.14 no.1
• /
• pp.1-10
• /
• 2013

This is a written version of the keynote speech delivered at the International Symposium on Hypersonic Aerothermodynamics - Recent Advances held in Bangalore, India, from December $6^{th}$ to $10^{th}$, 2012. In this document, what was accomplished in the past, the present status, and what is expected in the future in the field of hypersonic aerothermodynamics are reviewed. Solved problems are categorized into four items; unsolved problems into twelve items, and emerging problems into four items. Removing one degree uncertainty in trim angle of attack, studying the thermochemical phenomena in a hydrogen-helium-methane mixture, and entry flights of meteoroids are cited as the tasks for the future.

• Journal of the Korea Institute of Military Science and Technology
• /
• v.16 no.2
• /
• pp.175-184
• /
• 2013

In counterfire warfare, it is important to detect and attack enemy targets faster than the enemy using sensing and shooting assets. The artillery assets of North Korea are mostly mine artillery and much more than those of South Korea. To cope with sudden fire attacks from North Korea, we need to improve capability of our artillery. In this paper, we discuss issues and problems of our counterfire warfare systems and processes to overcome numerical inferiority and defend against the mine artillery. We develop a simulation model for counterfire warfare and analyze effectiveness of our counterfire weapon systems and operations based on various counterfire warfare scenarios. Finally, we propose methods of tactical operations and acquisition strategies of counterfire weapon systems, including detaction, firing, and protection assets.

• ETRI Journal
• /
• v.22 no.4
• /
• pp.25-31
• /
• 2000

Design of secure and efficient public-key encryption schemes under weaker computational assumptions has been regarded as an important and challenging task. As far as ElGamal-type encryption schemes are concerned, some variants of the original ElGamal encryption scheme based on weaker computational assumption have been proposed: Although security of the ElGamal variant of Fujisaki-Okamoto public -key encryption scheme and Cramer and Shoup's encryption scheme is based on the Decisional Diffie-Hellman Assumption (DDH-A), security of the recent Pointcheval's ElGamal encryption variant is based on the Computational Diffie-Hellman Assumption (CDH-A), which is known to be weaker than DDH-A. In this paper, we propose new ElGamal encryption variants whose security is based on CDH-A and the Elliptic Curve Computational Diffie-Hellman Assumption (EC-CDH-A). Also, we show that the proposed variants are secure against the adaptive chosen-ciphertext attack in the random oracle model. An important feature of the proposed variants is length-efficiency which provides shorter ciphertexts than those of other schemes.

• 제어로봇시스템학회:학술대회논문집
• /
• 2003.10a
• /
• pp.882-885
• /
• 2003

In this paper, we describe a haptic jacket and wheel as a haptic interface to enhance VR game realism. Building upon the VR game system using this devices, our haptic interface technique allows the user to intuitive interact on game contents, and then to sense the game event properties such as walking, attacking, driving and fire in a natural way. In addition, we extended the initial haptic model to support haptic decoration and dynamic interactions due to the added game event in a real time display. An application example presented here is a VR Dino-Attack game. This game supports interactions among dynamic and our intuitive haptic interface. Modeling physic interactions involves precise collision detection, real-time force computation, and high control-loop bandwidth.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.7 no.6
• /
• pp.1480-1491
• /
• 2013

Certificate-based cryptography is a new cryptographic primitive which eliminates the necessity of certificates in the traditional public key cryptography and simultaneously overcomes the inherent key escrow problem suffered in identity-based cryptography. However, to the best of our knowledge, all existed constructions of certificate-based encryption so far have to be based on the bilinear pairings. The pairing calculation is perceived to be expensive compared with normal operations such as modular exponentiations in finite fields. The costly pairing computation prevents it from wide application, especially for the computation limited wireless sensor networks. In order to improve efficiency, we propose a new certificate-based encryption scheme that does not depend on the pairing computation. Based on the decision Diffie-Hellman problem assumption, the scheme's security is proved to be against the chosen ciphertext attack in the random oracle. Performance comparisons show that our scheme outperforms the existing schemes.