Earticle

In this paper, a new approach based on the Honey bee mating optimization (HBMO) technique is proposed to tune the parameters of the multi-machine power system stabilizers (PSSs). The honey- bee mating process has been considered as a typical swarm-based approach to optimization, in which the search algorithm is inspired by the process of real honey-bee mating. The PSSs parameters tuning problem is converted to an optimization problem with time domain- based objective function which is solved by a HBMO algorithm. To ensure the robustness of the proposed stabilizers, the design process takes a wide range of operating conditions into account. The performance of the newly designed PSSs is evaluated in a three -machine power system subjected to the different types of operating conditions in comparison with the genetic algorithm based PSSs. The effectiveness of the proposed technique is demonstrated through nonlinear time-domain simulation studies over a wide range of loading condition.

In hierarchical network where a macro cell includes multiple small scale cells such as micro, femto cells and Wi-Fi hot-spots, preventing unnecessary handover is one of the most important matters due to so many inter-tier handovers. From the observation that there are many temporary visitors who stay in a small scale cell with short residence time, some recent handover schemes propose to delay the handover from macro to small scale cell to prevent unnecessary handovers made by those temporary visitors. Using such a conservative admission policy, we can prevent temporary visitor being handed over to small cell so that the number of inter-tier handovers can be significantly decreased. As a new concern of performance, we analyze the negative side effect of delayed handover scheme by using the theory of probability. Based on the analysis, we present some insights to determine the proper delay parameter to minimize the negative side effect probability.

Modeling of domain-dependent aspects is a key prerequisite for the design of software for mobile applications. Most mobile applications include a more or less advanced model of selected aspects of the domain in which they are used. This paper discusses the creation of such a model and its relevance for technical design of a mobile application. Conventional approaches to modeling of context focus either on the application domain or the problem domain. These approaches are presented and their relevance for technical design of software for mobile systems is discussed. The paper also discuss from an empirical study where a methodology that combines both of these approaches was introduced and employed for modeling of the domain-dependent aspects that were relevant for the design of a software component in a mobile applications. The empirical study was conducted in two companies that produce software for mobile applications. The resulting models of domain-dependent aspects are presented, and the experiences from the modeling process are discussed. It is concluded that a dual perspective based on both of the conventional approaches is relevant for capturing the aspects that are necessary for creating the domain-dependent models that are integrated in a mobile applications system.

In the routing processing of Internet of things, network traffic distribution changes constantly, network links and nodes will also fail randomly or some new nodes will be added; the autocatalytic and positive feedback mechanism of the ant colony algorithm conforms to the characteristics of the route searching. This paper uses ant colony algorithm to search route; uses the broadcast signaling, which is featured with the random multi-sending and the short life cycle, to overcome the problem of more network nodes and more variable network structure. The simulation results show that searching route by ant colony algorithm in Internet of things can reduce the broadcast storm effectively. With the increasing number of nodes in the route searching, the time of route setup was significantly shortened.

As network equipment consumes much more energy than ever before, research effort has been focusing on developing the access network protocol that aims to achieve communication with maximum energy efficiency. Ethernet is the most widely deployed access network protocol around the world. To improve energy efficiency in Ethernet protocol, IEEE 802.3az working group has defined low-power idle(LPI)-based Energy Efficient Ethernet (EEE) specification. In this paper, an enhanced synchronizing packet coalescing mechanism is proposed which improves energy efficiency in small Ethernet switch with basis on EEE specification. This mechanism is designed to predict traffic characteristic for active period by measuring incoming traffic amount during certain period. Once the traffic characteristic is predicted, threshold value would be adjusted to be optimal for the traffic load. Network performance was evaluated through simulation experiment that was performed assuming Poisson distribution traffic and burst traffic pattern respectively. Performance evaluation results indicate that the proposed mechanism improves overall performance compared to traditional mechanism, since it significantly reduces energy consumption rate, even though average packet delay increases a little bit.

This paper describes secure control architecture for UAV control system for multiple operator environments. Multiple access of UAV is necessary because UAV must capture and process various data and UAV's functions become more complex and diverse. From this perspective, UAV control system needs to be secure because new control system will be exposed to multiple operators. To discuss this issue, the paper proposes new control archi-tecture for UAV using RBAC (Roll Based Access Control). Generally, RBAC is applied and contributes to the security of computer systems, thus, we modified RBAC to apply to UAV control system. In this paper, we present how the modified RBAC model can be applied to UAV control systems and implement a prototype UAV control system based on the modified RBAC model.

In this paper, we propose an electronic ordering and payment system with embedded devices to improve a conventional manned ordering and payment method using a POS system for quick and easy ordering and payment. Embedded devices as thin clients of the proposed system can be placed on each table, and a user can choose a menu and can check his payment from the screen easily. After a user completes an order, the ordering message is transferred to server for preparing the chosen menu at a shop counter or a cooking place in a restaurant. The proposed system also provides various multimedia contents such as advertising information to users during wait time. Finally we successfully verified the feasibility and the effectiveness of the proposed system by conducting an experiment of ordering and payment checking with several embedded client systems connected to a server system. The proposed system can be applied to an unmanned ticketing or dispenser system, an information kiosk and a guidance robot.