• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.15 no.6
• /
• pp.1251-1255
• /
• 2011

In this paper, we propose an optimal path search algorithm that contains all nodes using genetic algorithm. An object in this approach is formed as an equation related with the Euclidean distance between an intermediate node and the starting node and between an intermediate node and the goal node. Like other genetic algorithm structures, our algorithm defines a fitness function and selects a crossover spot node and a bitwise crossover point. A new node out of such operation survives only if it satisfies the fitness criteria and that node then becomes the starting node for the next generation. Repetition continues until no changes are made in the population. The efficiency of this proposed approach is verified in the experiment that it is better than two other contestants - sequential approach and the random approach.

• Journal of Korean Society of Steel Construction
• /
• v.22 no.1
• /
• pp.33-42
• /
• 2010

Patterning with a geodesic line is essential for economical or efficient usage of membrane materialsin fabric tension membrane structural engineering and analysis. The numerical algorithm to determine the geodesic line for membrane structures is generally classified into two. The first algorithm finds a non-linear shape using a fictitious geodesic element with an initial pre-stress, and the other algorithm is the geodesic line cutting or searching algorithm for arbitrarily curved 3D surface shapes. These two algorithms are still being used only for the three-node plane stress membrane element, and not for the four-node element. The lack of a numerical algorithm for geodesic lines with four-node membrane elements is the main reason for the infrequent use of the four-node membrane element in membrane structural engineering and design. In this paper, a modified numerical algorithm is proposed for the generation of a geodesic line that can be applied to three- or four-node elements at the same time. The explicit non-linear static Dynamic Relaxation Method (DRM) was applied to the non-linear geodesic shape-finding analysis by introducing the fictitiously tensioned 'strings' along the desired seams with the three- or four-node membrane element. The proposed algorithm was used for the numerical example for the non-linear geodesic shape-finding and patterning analysis to demonstrate the accuracy and efficiency, and thus, the potential, of the algorithm. The proposed geodesic shape-finding algorithm may improve the applicability of the four-node membrane element for membrane structural engineering and design analysis simultaneously in terms of the shape-finding analysis, the stress analysis, and the patterning analysis.

• 한국정보디스플레이학회:학술대회논문집
• /
• 2006.08a
• /
• pp.739-742
• /
• 2006

In this paper, we propose a method for automatic generation of equivalent circuit for unit cell of LCDTV. In order to extract a circuit model, computer program generates electrical connectivity of resistors and capacitors from the layout through pattern analysis with electrode and port information. For combining two types of independent equivalent circuits, we propose a node insertion algorithm. As a consequence, we can generate an equivalent RC circuit without increasing the capacitive elements.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• v.9 no.2
• /
• pp.275-279
• /
• 2005

In this paper, we analyses of a traffic for authentication signaling in third generation mobile sensor network. In universal mobile telecommunication system, authentication functions are utilized to identify and authentication a mobile station and validate the service request network services. The authenticating parties are the authentication the serving general packet radio service support node access the authentication center to obtain the authentication with the mobile station. In this paper, we propose that the automatic cost-effective solution size of the authentication vector array.

• Proceedings of the KIEE Conference
• /
• 2002.07c
• /
• pp.1953-1955
• /
• 2002

An efficient automatic mesh generator suitable for the harsh model, which contains abrupt regions and thin layers, is implemented. It adopts two kinds of node placement scheme. In the beginning, the initial nodes on the boundary are generated automatically by an efficient boundary node generation scheme. And then the inner nodes are placed using nodal spacing value technique repeatedly from the initial mesh to final mesh. The effectiveness of the proposed technique is verified by a thin film micro-strip line modeling, where the final mesh contains no sliver elements.

• Proceedings of the Korean Operations and Management Science Society Conference
• /
• 2002.05a
• /
• pp.402-408
• /
• 2002

Multicast technology allows the transmission of data from one source node to a selected group of destination nodes. Multicast routes typically use trees, called multicast routing trees, to minimize resource usage such as cost and bandwidth by sharing links. Moreover, the quality of service (QoS) is satisfied by distributing data along a path haying no more than a given number of arcs between the root node of a session and a terminal node of it in the routing tree. Thus, a multicast routing tree for a session can be represented as a hop constrained Steiner tree. In this paper, we consider the hop-constrained multicast route packing problem with bandwidth reservation. Given a set of multicast sessions, each of which has a hop limit constraint and a required bandwidth, the problem is to determine a set of multicast routing trees in an arc-capacitated network to minimize cost. We propose an integer programming formulation of the problem and an algorithm to solve it. An efficient column generation technique to solve the linear programming relaxation is proposed, and a modified cover inequality is used to strengthen the integer programming formulation.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.22 no.7
• /
• pp.943-948
• /
• 2018

In this paper, we propose a hybrid approach of combining $A^*$ and Genetic algorithm in the path search problem. In $A^*$, the cost from a start node to the intermediate node is optimized in principle but the path from that intermediate node to the goal node is generated and tested based on the cumulated cost and the next node in a priority queue is chosen to be tested. In that process, we adopt the genetic algorithm principle in that the group of nodes to generate the next node from an intermediate node is tested by its fitness function. Top two nodes are selected to use crossover or mutation operation to generate the next generation. If generated nodes are qualified, those nodes are inserted to the priority queue. The proposed method is compared with the original sequential selection and the random selection of the next searching path in $A^*$ algorithm and the result verifies the superiority of the proposed method.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.23 no.5
• /
• pp.596-603
• /
• 2019

A reprogramming operation is necessary to update the software code of the node to change or update the functionality of the deployed node in wireless sensor networks. This paper proposes an optimization algorithm that minimizes the transmission energy of a node for the purpose of reprogramming a node in wireless sensor networks. We also design an algorithm that keeps energy consumption of all nodes balanced in order to maintain the lifetime of the network. In this paper, we propose a Tabu search algorithm with a new neighborhood generation method for minimizing transmission energy and energy consumption in wireless sensor networks with many nodes. The proposed algorithm is designed to obtain optimal results within a reasonable execution time. The performance of the proposed Tabu search algorithm was evaluated in terms of the node's transmission energy, remaining energy, and algorithm execution time. The performance evaluation results showed better performance than the previous methods.

• Journal of Genetic Medicine
• /
• v.20 no.2
• /
• pp.52-59
• /
• 2023

Purpose: Colorectal cancer (CRC) is a common malignancy worldwide and the second leading cause of cancer-related deaths. In addition, lymph node metastasis in CRC is considered an important prognostic factor for predicting disease recurrence and patient survival. Recent studies demonstrated that the microbiome makes substantial contributions to tumor progression, however, there is still unknown about the microbiome associated with lymph node metastasis of CRC. Here, we first reported the microbial and tumor-infiltrating immune cell differences in CRC according to the lymph node metastasis status. Materials and Methods: Using Next Generation Sequencing data acquired from 368 individuals diagnosed with CRC (N0, 266; N1, 102), we applied the LEfSe to elucidate microbial differences. Subsequent utilization of the Kaplan-Meier survival analysis enabled the identification of particular genera exerting significant influence on patient survival outcomes. Results: We found 18 genera in the N1 group and 3 genera in the N0 group according to CRC lymph node metastasis stages. In addition, we found that the genera Crenobacter (P=0.046), Maricaulis (P=0.093), and Arsenicicoccus (P=0.035) in the N0 group and Cecembia (P=0.08) and Asanoa (P=0.088) in the N1 group were significantly associated with patient survival according to CRC lymph node metastasis stages. Further, Cecembia is highly correlated to tumor-infiltrating immune cells in lymph node metastasized CRC. Concolusion: Our study highlights that tumor-infiltrating immune cells and intratumoral microbe diversity are associated with CRC. Also, this potential microbiome-based oncology diagnostic tool warrants further exploration.

• Proceedings of the KIEE Conference
• /
• 2006.07a
• /
• pp.169-170
• /
• 2006

In this paper, A new algorithm for load flow calculation is proposed for radial distribution network. Feeder Remote Terminal Unit (FRTU) is utilized to collect data such as current magnitude and angle of power factor at each node. Proposed algorithm is based on the model of distributed load in distribution system. Load flow calculation is using four terminal constants method.