• Proceedings of the Korean Society for Noise and Vibration Engineering Conference
• /
• 2008.11a
• /
• pp.668-669
• /
• 2008

In this paper, we introduce various coarse-grained elastic network modeling (ENM) techniques as a novel computational method for simulating atomic scale dynamics in macromolecules including DNA, RNA, protein, and polymer. In ENM, a system is modeled as a spring network among representative atoms in which each linear elastic spring is well designed to replace both bonded and nonbonded interactions among atoms in the sense of quantum mechanics. Based on this simplified system, a harmonic Hookean potential is defined and used for not only calculating intrinsic vibration modes of a given system, but also predicting its anharmonic conformational change, both of which are strongly related with its functional features. Various nano and bio applications of ENM such as fracture mechanics of nanocomposite and protein dynamics show that ENM is one of promising tools for simulating atomic scale dynamics in a more effective and efficient way comparing to the traditional molecular dynamics simulation.

• Proceedings of the Korean Operations and Management Science Society Conference
• /
• 1996.10a
• /
• pp.157-161
• /
• 1996

This paper presents a method for designing private networks considering response time under packet switching network environment Specifically, we propose the designing process of a private network, where the host is assumed to be synchronous and the terminals to be asynchronous, according to protocol and access lines. To calculate the response time for the networks, we use the mutual independent M/M/1 and M/W/s queueing models. Comparing the response time derived from the model with the actual one, it is found that the model yields almost same performance in a more economic way. From the experiment, it is suggested that the necessary number of lines as well as the response time for a specific network can be estimated properly by sensitivity analysis.

• Proceedings of the KIEE Conference
• /
• 2004.07d
• /
• pp.2528-2530
• /
• 2004

Neural network minimization problems are often conditioned and in this contribution way to handle this will be discussed. It is shown that a better conditioned minimization problem can be obtained if the problem is separated with respect to the linear parameters. This will increase the convergence speed of the minimization. One of the most powerful uses of neural networks is in function approximation(curve fitting)[1]. A main characteristic of this solution is that function (f) to be approximated is given not explicitly but implicitly through a set of input-output pairs, named as training set, that can be easily obtained from calibration data of the measurement system. In this context, the usage of Neural Network(NN) techniques for modeling the systems behavior can provide lower interpolation errors when compared with classical methods like polynomial interpolation. This paper solve of single-variable minimization using neural network.

• Journal of Ocean Engineering and Technology
• /
• v.12 no.2 s.28
• /
• pp.139-146
• /
• 1998

This paper introduces a network system for monitoring coastal oceanographic data. The network system consists of three parts such as the buoy to observe oceanographic data, the local site to collect data transferred from buoys, and the host site to construct the oceanographic data-base and to share the information for monitoring coastal oceanographic data. The buoy has a one-board microcomputer to manage and to acquire coastal environment data in real-time. A wireless and wire communication technique is employed in order to transfer data measured by buoys and to link local and host sites, respectively. In measuring coastal environment data, this system shows more cost-effective way than the presents conventional. In addition, the realtime monitoring system continuously from various sites with the network systems.

• 제어로봇시스템학회:학술대회논문집
• /
• 1991.10a
• /
• pp.147-151
• /
• 1991

In this paper a time-optimal path planning scheme for the multiple robot manipulators will be proposed by using hopfield neural network. The time-optimal path planning, which can allow multiple robot system to perform the demanded tasks with a minimum execution time and collision avoidance, may be of consequence to improve the productivity. But most of the methods proposed till now suffers from a significant computational burden and thus limits the on-line application. One way to avoid such a difficulty is to rearrange the problem as MTSP(Multiple Travelling Salesmen Problem) and then apply the Hopfield network technique, which can allow the parallel computation, to the minimum time problem. This paper proposes an approach for solving the time-optimal path planning of the multiple robots by using Hopfield neural network. The effectiveness of the proposed method is demonstrated by computer simulation.

• Proceedings of the KSR Conference
• /
• 2010.06a
• /
• pp.850-856
• /
• 2010

With skyrocketing number of people who use smartphone, the technology of wireless internet that is mobile-phone, Wi-Fi, Wibro, HSDPA has been developed remarkably. The environment that one can use wireless internet easily and inexpensively anywhere has been made. So, the customer of railway have needs and desire that could use the wireless internet network as travelling in rolling-stock. Although KTX, KTX-II and some subway agency has built-up wireless internet network and serviced it, it's closed infra and inconvenient environment has been criticized. So, we survey the wireless internet network that is serviced at present and the way to build wirelee internet infra to satisfy the needs of customers.

• Proceedings of the KIEE Conference
• /
• 1990.07a
• /
• pp.485-491
• /
• 1990

We propose a minimum-time path planning soheme for the robot manipulator using Hopfield neural network. The minimum-time path planning, which can allow the robot system to perform the demanded tasks with a minimum execution time, may be of consequence to improve the productivity. But most of the methods proposed till now suffers from a significant computational burden and thus limits the on-line application. One way to avoid such a difficulty is to apply the neural network technique, which can allow the parallel computation, to the minimum-time problem. This paper propose an approach for solving the minimum-time path planning by using Hopfield neural network. The effectiveness of the proposed method is demonstrarted using the PUMA 560 manipulator.

• International Journal of Contents
• /
• v.5 no.4
• /
• pp.88-93
• /
• 2009

It is important for communication networks to possess the capability to overcome failures and provide survivable services. We address modeling and analysis of performability affected by both performance and availability of system components for a token ring network under failure and repair conditions. Stochastic reward nets (SRN) is an extension of stochastic Petri nets and provides compact modeling facilities for system analysis. In this paper, hierarchical SRN modeling techniques are used to overcome state largeness problem. The upper level model is used to compute availability and the lower level model captures the performance. And Normalized Throughput Loss (NTL) is obtained for the composite ring network for each node failures occurrence as a performability measure. One of the key contributions of this paper constitutes the Petri nets modeling techniques instead of complicate numerical analysis of Markov chains and easy way of performability analysis for a token ring network under SRN reward concepts.

• Journal of computational fluids engineering
• /
• v.14 no.4
• /
• pp.86-92
• /
• 2009

Nowadays, the development of microfluidic chip [i.e. biochip, micro-total analysis system ($\mu$-TAS) and LOC (lab-on-a-chip)] becomes more active, and the microchannels to deliver fluid by pressure or electroosmotic forces tend to be more complex like electronic circuits or networks. For a simple network of channels, we may calculate the pressure and the flow rate easily by using suitable formula. However, for complex network it is not handy to obtain such information with that simple way. For this reason, Graphic User Interface (GUI) program which can rapidly give required information should be necessary for microchip designers. In this paper, we present a GUI program developed in our laboratory and the simple theoretical formula used in the program. We applied our program to simple case and could get results compared well with other numerical results. Further, we applied our program to several complex cases and obtained reasonable results.

• Journal of the Korea Society of Computer and Information
• /
• v.23 no.11
• /
• pp.95-101
• /
• 2018

In this paper, we propose an efficient friendly forces location data sharing algorithm in a troops using a low bandwidth radio. The future battlefield is a 'Network Centric Warfare' with a concept of identifying the position and power of the enemy and friendly forces and leading the battlefield to victory through proper links at the time of our need. One of the basic elements in the 'Network Centric Warfare' is to share friendly forces location data. The bandwidth and transmission rates of radio used in battalion are low. Nevertheless, we should share our locations data almost in real time for effective fighting in a war situation. This paper describes the efficient method of friendly forces location data sharing based on low bandwidth radio. In particular, the concept of 'network-centered warfare' is reflected in the troop below the battalion to present an integrated and efficient way to shared location data of friendly forces.