• The KIPS Transactions:PartC
• /
• v.14C no.2
• /
• pp.179-184
• /
• 2007

The problem of selecting base station location in the design of mobile communication system has been basically regarded as a problem of assigning maximum users in the cell to the minimum base stations while maintaining minimum SIR. and it is NP hard. The objective function of warehouse location problem, which has been used by many researchers, is not proper function in the base station location problem in CDMA mobile communication, The optimal and approximate solutions have been presented by using proposed object function and algorithms of exact solution, and the simulation results have been assessed and analyzed. The optimal and approximate solutions are found by using mixed integer programming instead of meta-heuristic search methods.

• Journal of the Korea Society of Computer and Information
• /
• v.19 no.4
• /
• pp.91-99
• /
• 2014

TThis paper suggests an heuristic polynomial time algorithm to solve the optimal solution for QAP (quadratic assignment problem). While Hungarian algorithm is most commonly used for a linear assignment, there is no polynomial time algorithm for the QAP. The proposed algorithm derives a grid type layout among unit distances of a distance matrix. And, we apply max-flow/min-distance approach to assign this grid type layout in such a descending order way that the largest flow is matched to the smallest unit distance from flow matrix. Evidences from implementation results of the proposed algorithm on various numerical grid type QAP examples show that a solution to the QAP could be obtained by a polynomial algorithm.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.27 no.5C
• /
• pp.422-434
• /
• 2002

We propose 3-layered hierarchical fiber-optic backbone transmission network composed of B-DCS, Backbone ring, Edge ring for efficient transmission of high capacity traffic and consider design method to ensure network survivability of each layer at affordable cost. Mathematical ring-construction cost minimization using MIP(Mixed Integer Programming) models results in NP-complete problem. So, it is hard to solve it within reasonable computing time. on a large-scale network. Therefore we develop heuristic algorithms solving WSCAP(Working and Spared Channel Assignment Problem) for B-DCS, MRLB(Multi-Ring Load Balancing) problem for Backbone ring, and ORLB(Overlayed Ring Load Balancing) problem for Edge ring and show their usefulness through case study.

• Journal of the Korea Computer Industry Society
• /
• v.3 no.2
• /
• pp.185-190
• /
• 2002

As the TSP(Traveling Salesman Problem) belongs to the class of NP-complete problems, various techniques are required for finding optimum or near optimum solution to the TSP. This paper designs a distributed genetic algorithm in order to reduce the execution time and obtain more near optimal using multi-slave model for the TSP. Especially, distributed genetic algorithms with multiple populations are difficult to configure because they are controlled by many parameters that affect their efficiency and accuracy. Among other things, one must decide the number and the size of the populations (demes), the rate of migration, the frequency of migrations, and the destination of the migrants. In this paper, I develop random dynamic migration rate that controls the size and the frequency of migrations. In addition to this, I design new migration policy that selects the destination of the migrants among the slaves

• Journal of the Korea Society of Computer and Information
• /
• v.20 no.5
• /
• pp.31-39
• /
• 2015

The degree-constrained minimum spanning tree (d-MST) problem is considered NP-complete for no exact solution-yielding polynomial algorithm has been proposed to. One thus has to resort to an heuristic approximate algorithm to obtain an optimal solution to this problem. This paper therefore presents a polynomial time algorithm which obtains an intial solution to the d-MST with the help of Kruskal's algorithm and performs k-opt on the initial solution obtained so as to derive the final optimal solution. When tested on 4 graphs, the algorithm has successfully obtained the optimal solutions.

• Journal of the Computational Structural Engineering Institute of Korea
• /
• v.14 no.3
• /
• pp.299-308
• /
• 2001

This paper presents the three-dimensional (3-D) study of the natural vibration of cantilevered laminated composite plates. The Ritz method is used to obtain stationary values of the associated Lagrangian functional with displacements approximated by mathematically complete polynomials satisfying the boundary conditions at the clamped edge exactly. The accuracy of the 3-D model is established through a convergence study of non-dimensional frequencies followed by a comparison of the converged 3-D solutions with analytical and experimental findings in the existing literature. A wide scope of 3-D frequency results explain the influence of a number of geometrical and material parameters for cantilevered laminated plates, namely aspect ratio (a/b), width-to-thickness ratio (a/h), orthotropy of material, number of plies (NP), fiber orientation angle(θ), and stacking sequence.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.31 no.11A
• /
• pp.1037-1043
• /
• 2006

This work presents an effective overlay multicast tree construction algorithm for synchronized real-time media multicast service over the Internet. The proposed algorithm is designed to minimize delay variance among group members to provide the synchronized service as well as average delay of group members in order to support the service in real-time. Basically, of orthogonal genetic algorithm is employed to obtain the near optimal tree with a low computational complexity since the given problem is NP-complete. Finally, experimental results are provided to show the superior performance of the proposed algorithm.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.21 no.2
• /
• pp.169-175
• /
• 2021

The lot sizing problem(LSP) is a hard problem that classified as non-deterministic(NP)-complete because of the polynomial-time optimal solution algorithm is unknown yet. The well-known W-W algorithm can be obtain the solution within polynomial-time, but this algorithm is a very complex, therefore the heuristic approximated S-M algorithm is suggested. This paper suggests O(n) linear-time complexity algorithm that can be find not the approximated but optimal solution. This algorithm determines the lot size Xt∗ in period t to the sum of the demands of interval [t,t+k], the period t+k is determined by the holding cost will not exceed setup cost of t+k period. As a result of various experimental data, this algorithm finds the optimal solution about whole data.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.24 no.3
• /
• pp.103-108
• /
• 2024

Kakuro puzzles are NP-complete problems where no way to solve puzzles in polynomial time is known. Until now, a brute-force search method or a linear programming method has been applied to substitute all possible cases. This paper finds a magic rule, a rule for box sizes and unfilled numbers according to sum clues. Based on the magic rule, numbers that cannot enter empty cells were deleted from the box for row and column sum clues. Next, numbers that cannot enter the box were deleted based on the sum clue value. Finally, cells with only a single number were confirmed as clues. As a result of applying the proposed algorithm to seven benchmarking experimental data, it was shown that solutions could be obtained for all problems.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.15 no.1
• /
• pp.207-215
• /
• 2015

This paper proposes a solution-yielding linear time algorithm to NP-complete Sudoku, to which no polynomial time algorithm has been proposed. The proposed algorithm is performed on blocks in the descending order of the number of clues they contain. It firstly determines all numbers that could possibly occur in the blank rows and columns of each block. By deriving an intersecting value of corresponding rows and columns, it assigns the final number for each blank. When tested on the traditional $9{\times}9$ Sudoku, the proposed algorithm has succeeded in obtaining the solution through performance of 9 times, the exact number of the blocks. Test results on modified Jigsaw Sudoku (9 blocks) and Hypersudoku (13 blocks) also show its success in deriving the solutions by execuring 9 and 13 times respectively. Accordingly, this paper proves that the Sudoku problem is in fact P-problem.