• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.15 no.3
• /
• pp.1100-1118
• /
• 2021

In a wide range of ML applications, the training data contains privacy-sensitive information that should be kept secure. Training the ML systems by privacy-sensitive data makes the ML model inherent to the data. As the structure of the model has been fine-tuned by training data, the model can be abused for accessing the data by the estimation in a reverse process called model inversion attack (MIA). Although, MIA has been applied to shallow neural network models of recognizers in literature and its threat in privacy violation has been approved, in the case of a deep learning (DL) model, its efficiency was under question. It was due to the complexity of a DL model structure, big number of DL model parameters, the huge size of training data, big number of registered users to a DL model and thereof big number of class labels. This research work first analyses the possibility of MIA on a deep learning model of a recognition system, namely a face recognizer. Second, despite the conventional MIA under the white box scenario of having partial access to the users' non-sensitive information in addition to the model structure, the MIA is implemented on a deep face recognition system by just having the model structure and parameters but not any user information. In this aspect, it is under a semi-white box scenario or in other words a gray-box scenario. The experimental results in targeting five registered users of a CNN-based face recognition system approve the possibility of regeneration of users' face images even for a deep model by MIA under a gray box scenario. Although, for some images the evaluation recognition score is low and the generated images are not easily recognizable, but for some other images the score is high and facial features of the targeted identities are observable. The objective and subjective evaluations demonstrate that privacy cyber-attack by MIA on a deep recognition system not only is feasible but also is a serious threat with increasing alert state in the future as there is considerable potential for integration more advanced ML techniques to MIA.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.27 no.8C
• /
• pp.748-757
• /
• 2002

Cyber-terror trials are increased in nowadays and these attacks are commonly using security vulnerability and information gathering method by variable services grew by the continuous development of Internet Technology. IDS's application environment is affected by this increasing Cyber Terror. General Network based IDS detects intrusion by signature based Intrusion Detection module about inflowing packet through network devices. Up to now security in network is commonly secure host, an regional issue adopted in special security system but these system is vulnerable intrusion about the attack in globally connected Internet systems. Security mechanism should be produced to expand the security in whole networks. In this paper, we analyzer the DARPA's program and study Infusion Detection related Technology. We design policy security framework for policy enforcing in whole network and look at the modules's function. Enforcement of security policy is acted by Intrusion Detection system on gateway system which is located in network packet's inflow point. Additional security policy is operated on-line. We can design and execute central security policy in managed domain in this method.

• The Journal of Society for e-Business Studies
• /
• v.25 no.2
• /
• pp.99-108
• /
• 2020

Unlike a general IT environment, an industrial control system is an environment where stability and continuity are more important than security. In the event of a security accident in the industrial control system, physical motion can be controlled, so physical damage can occur and physical damage can even result in personal injury. Cyber attacks on industrial control systems are not simply cyber damage, but terrorism. However, the security of industrial control systems has not been strengthened yet, and many vulnerabilities are actually occurring. This paper shows that the PLC can be remotely controlled by analyzing the connection process and packets for the PLC protocol used in the industrial control system and bypassing the security mechanism existing in the protocol. Through this, we intend to raise the security awareness of the industrial control system.

• Journal of KIISE:Computer Systems and Theory
• /
• v.33 no.8
• /
• pp.524-535
• /
• 2006

With rapid development of Internet technology, business management in an organization or an enterprise depends on Internet-based technology for the most part. Furthermore, as dependency and cohesiveness of network in the communication facilities are increasing, cyber attacks have been increased against vulnerable resource in the information system. Hence, to protect private information and computer resource, research for damage propagation is required in this situation. However the proposed traditional models present just mechanism for risk management, or are able to be applied to the specified threats such as virus or worm. Therefore, we propose the probabilistic model of damage propagation based on the Markov process, which can be applied to diverse threats in the information systems. Using the proposed model in this paper, we can predict the occurrence probability and occurrence frequency for each threats in the entire system.

• Convergence Security Journal
• /
• v.13 no.6
• /
• pp.51-59
• /
• 2013

Cyberspace, based on the dramatic development of information and communications technology, has brought enormous benefits to mankind. However, concerns over cyber terrorism and cyber attack are becoming serious. It is time to expand the global dialogue on international security issues in cyberspace. It is imperative to have a common understanding that cyberspace, the infrastructure for prosperity, should not be utilized as a space to create conflicts among states, and that all states agree to build confidence and peace in cyberspace. For this purpose, there are 3 tracks of international cooperations: 1)international cooperation such as UN and Conference on Cyberspace, 2)regional cooperations such as ARF and OSCE. 3)bilateral cooperations such US-Russia Cybersecurity Agreement, US-China presidential level dialogue. This paper will analyze the 1st track of international cooperations of UN and Conference on Cyberspace. With this, Korean government can prepare the forthcoming GGE activities and make our own strategy to deal with the global norms of good behaviour in cyberspace.

• Journal of the Korea Society for Simulation
• /
• v.15 no.1
• /
• pp.87-95
• /
• 2006

We need to check into when a security system is newly developed, we against cyber attack which is expected in real network. However it is impossible to check it under the environment of a large-scale distributive network. So it is need to simulate it under the virtual network environment. SSFNet is a event-driven simulator which can be represent a large-scale network. Unfortunately, it doesn't have the module to simulate security functions. In this paper, we added the IDS module to SSFNet. We implement the IDS module by modeling a key functions of Snort. In addition, we developed some useful functions using Java language which can manipulate easily a packet for network simulation. Finally, we performed the simulation to verify the function if our developed IDS and Packets Manipulation. The simulation shows that our expanded SSFNet can be used to further large-scale security system simulator.

• IEIE Transactions on Smart Processing and Computing
• /
• v.2 no.3
• /
• pp.168-175
• /
• 2013

Key escrow is a default property that is inherent in identity-based cryptography, where a curious private key generator (PKG) can derive a secret value shared by communicating entities in its domain. Therefore, a dishonest PKG can encrypt and decrypt ciphers or can carry out any attack on the communicating parties. Of course, the escrow property is not completely unwanted but is acceptable in other particular applications. On the other hand, in more civil applications, this key escrow property is undesirable and needs to be removed to provide maximum communication privacy. Therefore, this paper presents an escrow-free identity-based key agreement protocol that is also applicable even in a distinct PKG condition that does not use pairings. The proposed protocol has comparable computational and communicational performance to many other protocols with similar security attributes, of which their security is based on costly bilinear pairings. The protocol's notion was inspired by McCullagh et al. and Chen-Kudla, in regard to escrow-free and multi-PKG key agreement ideas. In particular, the scheme captures perfect forward secrecy and key compromise impersonation resilience, which were lacking in McCullagh et al.'s study, as well as all other desirable security attributes, such as known key secrecy, unknown key-share resilience and no-key control. The merit in the proposed protocol is the achievement of all required security requirements with a relatively lower computational overhead than many other protocols because it precludes pairings.

• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.29 no.6
• /
• pp.1403-1412
• /
• 2019

Effect of damage and loss cost for downtime is huge, if physical devices such as turbines, pipe, and storage tanks are in the abnormal state originated from not only aging, but also cyber attacks on the control and monitoring system like PLC (Programmable Logic Controller). To improve availability and dependability of the physical devices, we design and implement an indirect health monitoring system which sense temperature, acceleration, current, etc. indirectly, and put sensor data into Influx DB in real-time. Then, the actual performance of detecting abnormal state is shown using the indirect health monitoring system. Analyzing data are acquired using the real-time indirect health monitoring system, abnormal state and security threats can be double-monitored and lower maintenance cost utilizing prognostics and health management.

• Convergence Security Journal
• /
• v.21 no.1
• /
• pp.25-31
• /
• 2021

The purpose of this study was to keep the Security Control Center, which operates under a shift system, uninterrupted during the COVID-19 virus epidemic. Security facilities responding to cybersecurity threats are essential security facilities that must be operated 24 hours a day, 365 days a day in real time, and are critical to security operations and management. If security facilities such as infectious disease epidemic, system failure, and physical impact are closed or affected, they cannot respond to real-time cyberattacks and can be fatal to security issues. Recently, there have been cases in which security system facilities cannot be operated, such as the closure of facilities due to the COVID-19 virus epidemic and the availability of security systems due to the rainy season, and other cases need to be prepared. In this paper, we propose a plan to configure a security system facility as a multiplexing facility and operate it as an alternative in the event of a closed situation.

• Journal of Internet of Things and Convergence
• /
• v.7 no.3
• /
• pp.1-8
• /
• 2021

With the rapid increase in the use of IoT and mobile devices, cyber criminals targeting IoT devices are also on the rise. Among IoT devices, when using a wireless access point (AP), problems such as packets being exposed to the outside due to their own security vulnerabilities or easily infected with malicious codes such as bots, causing DDoS attack traffic, are being discovered. Therefore, in this study, in order to actively respond to cyber attacks targeting IoT devices that are rapidly increasing in recent years, we proposed a method to collect traces of intrusion incidents artifacts from IoT devices, and to improve the validity of intrusion analysis data. Specifically, we presented a method to acquire and analyze digital forensics artifacts in the compromised system after identifying the causes of vulnerabilities by reproducing the behavior of the sample IoT malware. Accordingly, it is expected that it will be possible to establish a system that can efficiently detect intrusion incidents on targeting large-scale IoT devices.