Hyunsu Jang

Adaptive Access Control Scheme Utilizing Context Awareness in Pervasive Computing Environments

In pervasive computing environments, where various types of information are publicly owned, and multiple users access the networks via various networked devices anytime and anywhere, access control that grants permission to an authorized user is definitely needed for secure information access. Context awareness refers to the idea that computers can both sense and react based on various context in their environments. In many access control schemes, recently, context awareness has been utilized to guarantee dynamic access control according to current context and various access control schemes utilizing context awareness have been proposed. However, previous studies have difficulty describing conditions for assigning roles and modifying permissions. They also simply consider assigning roles or modifying permissions, rather than providing detailed access control algorithms such as role delegation or role revocation. In this paper, we propose an adaptive access control scheme utilizing context awareness in pervasive computing environments. We design an adaptive access control model based on traditional RBAC(Role-Based Access Control) model, and present an adaptive access control scheme to guarantee dynamic user and permission assignment according to changes of context. In this scheme, we define context requirements in each table, enabling a more detailed description. We also guarantee dynamic access control via various access control algorithms.

A Survey on Interfaces to Network Security Functions in Network Virtualization

Network Functions Virtualization (NFV) opens new opportunities and challenges for security community. Unlike existing physical network infrastructure, in a virtualized network platform, security services can be dynamically deployed and maintained to cope with the threat of sophisticated network attacks that are increasing over time. This paper surveys the activity that many security vendors and Internet service providers are trying to define common interfaces for NFV-based security services through the analysis of use cases and related technologies. This activity is currently lead by Internet Engineering Task Force (IETF) that is an international Internet standardization organization.

The Analysis of Random Propagating Worms using Network Bandwidth

There is a well-defined propagation model, named the random constant spread (RCS) model, which explains worms that spread their clones with a random scanning strategy. This model uses the number of infected hosts in a domain as a factor in the worms’ propagation. However, there are difficulties in explaining the characteristics of new Internet worms because they have several considerable new features: the denial of service by network saturation, the utilization of a faster scanning strategy, a smaller size in the worm’s propagation packet, and to cause maximum damage before human-mediated responses are possible. Therefore, more effective factors are required instead of the number of infected hosts. In this paper, the network bandwidth usage rate is found to be an effective factor that explains the propagations of the new Internet worms with the random scanning strategy. The analysis and simulation results are presented using this factor. The simulation results show that the scan rate is more sensitive than the propagation packet for detecting worms’ propagations.

User Adaptive Recommendation Model by Using User Clustering based on Decision Tree

With the rapid growth of information and communication technology, many researchers are studying on development of user adaptive recommendation systems for user centric services. Most of the recommendation systems are being studied on using content-based and collaborative recommendation methods. However, these systems have the problems such as taking too much time for analyzing characteristics of new users or new services when they come into the system and generating too simple recommendation results due to the properties known as overspecialization and sparsity. In this paper, we propose an agent based recommendation model that can reduce analysis time when new users or new services appear in the system and recommend more user centric services. Proposed model clusters existing users by using decision tree and analyzes new incoming users by traversing the decision tree, which has already been constructed into the structure that reduces the analysis time. To prove the effectiveness of the proposed model, we implement user clustering and service recommendation scheme using decision tree, and evaluate its performance with some experimentations.

Analysis of the propagation pattern of a worm with random scanning strategy based on usage rate of network bandwidth

There have been many studies on modeling the propagation patterns of Internet worms since the advent of Morris worm. Among them, there is a well defined propagation model, which is generally called random constant spread (RCS) model. However, there are some limitations to model the propagation patterns of new emergent Internet worms with the RCS model because the model uses the only number of infected hosts as the factor of a worms propagation. The new worms have several considerable characteristics: utilization of a faster scanning strategy, miniaturization of the size of a worms propagation packet, denial of service by network saturation, and maximum damage before human-mediated responses. These characteristics make it difficult to notice much harder than before whether a worm propagates itself or not. Therefore, a basic factor instead of the number of infected hosts, which is used by the RCS model, is required to model the propagation patterns of new worms. In this paper, only analysis and simulation results based on usage rate of network bandwidth, which can be considered as a basic factor, are presented about the propagation pattern of a worm with random scanning strategy. Miniaturization of the size of a propagation packet and utilization of a faster scanning strategy are related to the size of worms propagation packet and its propagation rate, respectively. It is presented that the latter is more sensitive than the former.

Design of a Multi-middleware Bridge for Supporting Interoperability in Home Network Environments

We propose multi middleware bridge (MMB) scheme for supporting interoperability among several middlewares in home network environments. Our proposed bridge utilizes a bridge repository to support interoperability via the cooperation of bridge systems in the network. We examine the superiority of the proposed MMB scheme through the evaluation and comparison with existing bridge systems in the scope of qualitative perspective.

An Efficient Migration Scheme for Mobile Agents in Ubiquitous Environments

Along with the development of lots of mobile agent platforms, the various researches on migration schemes for mobile agents have been suggested. The agents play a key role to manage a huge amount of tasks instead of human being in ubiquitous computing environments. During their operations, agents may cause excessive overhead of network traffic because the size of agent code continuously becomes bigger than before as the increasing complexity of its functionalities. In addition, there is one additional possibility that frequent visits by a lot of agents make target platforms be the bottleneck of communication. To solve these problems, we suggest a new approach. The proposed scheme is to cache the codes of mobile agents which bases on frequent visiting of platforms in ubiquitous environments. To show superiority of our scheme, we compare the proposed method with the existing methods through various simulations.

Overlay Ring Based Secure Group Communication Scheme for Mobile Agents

Among the various inter-agent communication models proposed for multi-agent systems so far, several group communication schemes have been issued to guarantee the transparent communication among the agents. However, in mobile agent environments where each agent has mobility in the network, those schemes are not sufficient to fully handle the topological change caused by the migration of mobile agents. Also, these group communication schemes should be secure in order for them to be practical. In this paper, we propose a secure group communication scheme which applies the hierarchical overlay ring structure of mobile agents. The proposed scheme uses the ring channel in order to cope adaptively with the change of ring topology. The ring channel has the fundamental information to construct the ring and is managed by only the mobile agent platforms. Therefore, each mobile agent does not need to directly handle the ring channel and can perform group communication regardless of the changing ring topology.

AIM: Design and Implementation of Agent-based Intelligent Middleware for Ubiquitous HCI Environments

With the emergence of ubiquitous computing era, it has become increasingly important for a middleware which takes full advantage of HCI factors to support user-centric services. Many kinds of studies on HCI-friendly middleware for supporting user-centric services have been performed. However, previous studies have problems in supporting HCI factors, which are needed for user-centric services. In this paper, we present an agent-based intelligent middleware, which is called AIM, that provides user-centric services in ubiquitous HCI environments. We describe the middleware requirements for user-centric services by analyzing various HCI-friendly middleware and design AIM middleware which effectively supports various HCI factors such as context information management, pattern inference of user`s behavior, and dynamic agent generation, etc. We introduce service scenarios based on the user`s modalities in smart spaces. Finally, prototype implementation is illustrated as a manifestation of the benefits of the introduced infrastructure.

An Integrated Context Generation Scheme based on Ant Colony System

With the development of ubiquitous computing technology, the number of HCI applications is increasing, where they utilize various contexts to provide adaptive services to users according to the change of contexts, and also, technologies for collecting various sensor data and generating integrated contexts get more important. However, the research on the collection and integration of multi-sensor data is not sufficient when we consider the various utilization areas of the integrated contexts. In particular, they have some problems to be solved such as duplication of the context data and the high system load. In this paper, we propose an integrated context generation scheme based on Ant Colony System. Proposed scheme generates the context data as a form of XML and avoids the generation of unnecessary context information by detecting the repeated sensor information based on the ant colony system. As a result of detections, we reduce wasted resources and repositories when the integrated context is created. We also reduce the overhead for reasoning.