Nam-Uk Kim

Attribute Relationship Evaluation Methodology for Big Data Security

There has been an increasing interest in big data and big data security with the development of network technology and cloud computing. However, big data is not an entirely new technology but an extension of data mining. In this paper, we describe the background of big data, data mining and big data features, and propose attribute selection methodology for protecting the value of big data. Extracting valuable information is the main goal of analyzing big data which need to be protected. Therefore, relevance between attributes of a dataset is a very important element for big data analysis. We focus on two things. Firstly, attribute relevance in big data is a key element for extracting information. In this perspective, we studied on how to secure a big data through protecting valuable information inside. Secondly, it is impossible to protect all big data and its attributes. We consider big data as a single object which has its own attributes. We assume that a attribute which have a higher relevance is more important than other attributes.

Cyber military strategy for cyberspace superiority in cyber warfare

In this paper, we proposed robust and operational cyber military strategy for cyberspace superiority in cyber warfare. We considered cyber forces manpower, cyber intelligence capability, and an organization of cyber forces for improving cyber military strategy. In cyber forces power field, we should cultivated personals that can perform network computer operations and hold cyber security technology such as cyber intelligence collection, cyber-attack, cyber defence, and cyber forensics. Cyber intelligence capability includes cyber surveillance/reconnaissance, cyber order of battle, pre-CTO, and cyber damage assessment. An organization of cyber forces has to change tree structure to network structure and has to organize task or functional centric organizations. Our proposed cyber military strategy provides prior decision of action to operate desired effects in cyberspace.

Big data analysis system concept for detecting unknown attacks

Recently, threat of previously unknown cyber-attacks are increasing because existing security systems are not able to detect them. Past cyber-attacks had simple purposes of leaking personal information by attacking the PC or destroying the system. However, the goal of recent hacking attacks has changed from leaking information and destruction of services to attacking large-scale systems such as critical infrastructures and state agencies. In the other words, existing defence technologies to counter these attacks are based on pattern matching methods which are very limited. Because of this fact, in the event of new and previously unknown attacks, detection rate becomes very low and false negative increases. To defend against these unknown attacks, which cannot be detected with existing technology, we propose a new model based on big data analysis techniques that can extract information from a variety of sources to detect future attacks. We expect our model to be the basis of the future Advanced Persistent Threat(APT) detection and prevention system implementations.

Energy-Efficient Clustering scheme with Concentric Hierarchy

With the development of the Wireless Sensor Network (WSN) technology, ubiquitous technology comes to the fore as the core technology in the future. In the Wireless Sensor Network, energy efficiency of the whole network is a key problem that has to be solved. Clustering is one of routing techniques to improve energy efficiency. LEACH and LEACH-C are existing techniques focused on optimizing energy efficiency of the network by applying clustering. However, those protocols still have problems. LEACH does not guarantee balanced distribution of cluster heads and communication overhead of LEACH-C is quite heavy. In this paper, therefore, we suggest a new method, “Energy-Efficient Clustering scheme with Concentric Hierarchy (EECCH),” a centralized clustering scheme aimed at overcoming weaknesses of LEACH and LEACH-C both. By drawing circles with the base station as its center, the base station separates network nodes into some levels. Afterward, we let the clusters have different numbers of its member nodes to eliminate inequality in energy dissipation. Through this process, it becomes possible to improve energy efficiency. By using MATLAB, we prove that EECCH gives better performance than some other protocols which also use clustering technique, including LEACH, in the performance evaluation.

A Remote Control Architecture for Thin-Client in Mobile Cloud Computing

Smart mobile devices such as smartphone, and smart pad have rapidly pervaded our everyday life. If the limitation of their relatively poor computing resources such as processor, memory, battery can be expanded, smart mobile devices will become more powerful tool for human life. Mobile cloud computing, in which resource-rich virtual machines of smart mobile device are provided to a customer as a service, can be terrific solution for expanding the limitation of real smart mobile device. However, existing cloud computing technique for this kind of service is very poor to meet the attractive quality. Especially, remotely interacting with virtual machine (VM) smoothly as if the VM is the real device itself is the most challenging part. Existing remote control protocols for desktop can be suggested as an alternative, but they have some fatal flaws. In this paper, we propose the remote control architecture for real-time, close-knit interaction between real smart mobile device and virtual one.

An efficient hash-based load balancing scheme to support parallel NIDS

Today, as the scale of network grows up, a standalone NIDS with only one intrusion detection node is not enough to inspect all traffic. One of the most widely considered solutions to address this problem is to configure parallel NIDS in which multiple intrusion detection nodes work together. A load balancing mechanism enables this configuration by distributing traffic load to several nodes. In the frequently changing environment of todays network, it is an important issue for load balancing mechanism to distributing traffic equally to each node. Meanwhile, several studies have been made on the load balancing scheme, but they do not satisfy the requirements of load balancing for parallel NIDS. Thus we proposed HLPN (Hash-based Load balancing scheme suitable for Parallel NIDS) which satisfies these requirements. As a result of the performance evaluation, HLPN represented 58% better performance in terms of the fairness of the traffic distribution than static hash-based scheme, and gave almost equal, or rather better, performance to that of DHFV.

Adaptive keypoint-based CU depth decision for HEVC intra coding

While adaptive splitting the quadtree-structured coding units according to the video content is utmost important in achieving the best encoding performance of the High Efficiency Video Coding (MPEG-H HEVC/H.265), the newest video coding standard, its optimal decision requires much computation. To resolve this complexity problem, we extend the keypoint-based CU depth decision (KCD) method by further providing it with adaptive decision capability to reduce its complexity more. Experimental results show that the adaptive method can save encoding time by 11.6% with sacrificing BD-rate only by 0.68% compared to its non-adaptive scheme.

Applying Scheduling Algorithms with QoS in the Cloud Computing

Cloud computing is the model to use existing computing resources that are delivered as a form of service over a network. These services can be divided into three parts with software, platform, and infrastructure. It is important to evaluate cloud computing environment to predict valid cost to manage the cloud computing system. SimJava and GridSim is well-known simulation tools but they do not support the virtualization of cloud computing. CloudSim is only tool which can evaluate the performance of this environment and it is based on SimJava and GridSim. It is suitable to simulate the situation with large amount of devices and data in cloud computing. Also, it can simulate the virtualization of computing nodes, network devices, and storage units. Service provider has to guarantee quality of service to provide stable related services. For this, we can use the scheduling algorithms. However, there is no consideration of data priority in CloudSim. It is important to support QoS to keep the service level agreement. Thus, it is needed to research a scheduling algorithm to support QoS. In this paper, we propose the way to support various scheduling algorithms in CloudSim.

Improved CloudSim for Simulating QoS-Based Cloud Services

CloudSim is useful tool for a simulation of cloud environment. CloudSim can view the availability, a power consumption, and a network traffic of services on cloud environment. However, CloudSim cannot support Quality of Service (QoS). Thus, this paper proposes a cloud-simulating system with QoS using CloudSim. Providing CloudSim with a priority queue is basic solution for offering QoS. Additionally, this paper introduces implementation of priority queue in CloudSim. Proposed system is able to control cloudlets with priority and process them differentially. This advanced CloudSim shows faster complete time than default system in time-sharing policy.

Study on sensitive information leakage vulnerability modeling

Purpose – The purpose of this paper is to provide a model for quantitatively analyzing the security profile of an organization’s IT environment. The model considers the security risks associated with stored data, as well as services and devices that can act as channels for data leakages. The authors propose a sensitive information (SI) leakage vulnerability model. Design/methodology/approach – Factors identified as having an impact on the security profile are identified, and scores are assigned based on detailed criteria. These scores are utilized by mathematical models that produce a vulnerability index, which indicates the overall security vulnerability of the organization. In this chapter, the authors verify the model result extracted from SI leakage vulnerability weak index by applying the proposed model to an actual incident that occurred in South Korea in January 2014. Findings – The paper provides vulnerability result and vulnerability index. They are depends on SI state in information systems. Ori...