• The KIPS Transactions:PartA
• /
• v.13A no.1 s.98
• /
• pp.53-56
• /
• 2006

Genetic algorithms(GA), which are based on the idea of natural selection and natural genetics, have proven successful in solving difficult problems that are not easily solved through conventional methods. The classical GA has the problem to spend much time when population is large. Parallel genetic algorithm(PGA) is an extension of the classical GA. The important aspect in PGA is migration and GA operation. This paper presents PGAs that use fuzzy logic. Experimental results show that the proposed methods exhibit good performance compared to the classical method.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.20 no.3
• /
• pp.428-433
• /
• 2010

The notion "block rotation" denotes the operation of exchanging two consecutive sequences of elements uv to vu. There are three already well-known block rotation algorithms called BlockRotation, Juggling and Reversal algorithm. Recently we presented a novel block rotation algorithm called QuickRotation. In this paper we compare QuickRotation to these three known block rotation algorithms. This comparison covers a complexity analysis as well as benchmarking and shows that a switch to QuickRotation is almost always advantageous.

• Journal of KIISE:Computing Practices and Letters
• /
• v.16 no.4
• /
• pp.487-491
• /
• 2010

In this paper we propose a new single-query path planning algorithm for working well in high-dimensional configuration space. With the notice of the similarity between single-query algorithms with exploration algorithms, we propose a new path planning algorithm, which applies the Remembering Exploration method, which is one of exploration algorithms, to a path-planning problem by selecting a node from a roadmap, finding out the neighbor nodes from the node, and then inserting the neighbor nodes into the roadmap, recursively. For the performance comparison, we had experiments in 2D and 3D environments and compared the time to find out the path. In the results our algorithm shows the superior performance than other path planning algorithms.

• The Journal of The Korea Institute of Intelligent Transport Systems
• /
• v.12 no.6
• /
• pp.1-9
• /
• 2013

The studies on the shortest path algorithms based on Dijkstra algorithm has been done continuously to decrease the time for searching. $A^*$ algorithm is the most represented one. Although fast searching speed is the major point of $A^*$ algorithm, there are high rates of failing in search of the shortest path, because of complex and irregular networks. The failure of the search means that it either did not find the target node, or found the shortest path, witch is not true. This study proposed $A^*$ algorithm applying method that can reduce searching failure rates, preferentially organizing the relations between the starting node and the targeting node, and appling it in reverse according to the organized path. This proposed method may not build exactly the shortest path, but the entire failure in search of th path would not occur. Following the developed algorithm tested in a real complex networks, it revealed that this algorithm increases the amount of time than the usual $A^*$ algorithm, but the accuracy rates of the shortest paths built is very high.

• Journal of the Korean Society of Marine Environment & Safety
• /
• v.25 no.2
• /
• pp.169-177
• /
• 2019

It is very important to understand the intention of a target ship to prevent collisions in multiple-ship situations. However, considering the intentions of a large number of ships at the same time is a great burden for the officer who must establish a collision avoidance plan. With a distributed algorithm, a ship can exchange information with a large number of target ships and search for a safe course. In this paper, I have applied a Distributed Stochastic Search Algorithm (DSSA), a distributed algorithm, for ship collision avoidance. A ship chooses the course that offers the greatest cost reduction or keeps its current course according to probability and constraints. DSSA is divided into five types according to the probability and constraints mentioned. In this paper, the five types of DSSA are applied for ship collision avoidance, and the effects on ship collision avoidance are analyzed. In addition, I have investigated which DSSA type is most suitable for collision avoidance. The experimental results show that the DSSA-A and B schemes offered effective ship collision avoidance. This algorithm is expected to be applicable for ship collision avoidance in a distributed system.

• Journal of the Korea Computer Industry Society
• /
• v.5 no.9
• /
• pp.1079-1088
• /
• 2004

Generally, the code motion algorithm accomplishes the run-time optimal connected with the computational optimifation and the register overhead This paper proposes a sparse code motion, which considers the code size, in addition to computational optimization and lifetime optimization. The BCM algorithm carries out the optimal code motion computationally and the LCM algorithm reduces the register overhead in a sparse code motion algorithm. A sparse code motion algorithm is optimum algorithm computationally and lifetime because of suppression unnecessary code motion This algorithm improves runtime and efficiency of the program than the previous work through the performance test.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.27 no.6B
• /
• pp.629-637
• /
• 2002

RLS a1gorithm is a kind of the adaptive a1gorithms in smart antennas and adapts the weight vector using the difference between the output signal of array antennas and the known training sequence. In this paper, we propose a new algorithm based on the RLS algorithm. It calculates the error signal with reference signal derived from blind scheme. Simulation results show that the proposed algorithm yields more user capacity by 67∼74% than other blind adaptive algorithms(LS-DRMTA, LS-DRMTCMA) at the same BER and the beamformer forms null beams toward interference signals and the main beam toward desired signal.

• Proceedings of the Korean Society of Computer Information Conference
• /
• 2013.07a
• /
• pp.293-294
• /
• 2013

다양한 생물체의 행동 원리를 이용하여 모델링한 알고리즘을 생체모방 알고리즘(Biologically Inspired Algorithm)이라고 한다. 본 논문에서는 생체모방 알고리즘 중 동물 집단의 행동을 모델링한 플로킹 알고리즘(Flocking Algorithm)을 이용한 집단 게임 캐릭터의 길찾기 방법을 제안한다. 플로킹 알고리즘의 조타행동은 크게 분리(Separation), 정렬(Alignment), 응집(Cohesion), 회피(Avoidance)로 구성되어 있다. 게임에서의 기존 플로킹 알고리즘은 주로 여러 개의 몬스터나 NPC 들로 구성된 몇 개의 그룹 단위로 독자적인 집단 행동을 하는 경우에 적합하다. 그러나, 게임플레이어가 제어하는 캐릭터가 많은 경우, 기존 알고리즘은 플레이어가 선택한 캐릭터 그룹을 목표지점으로 이동하는 방법으로 적합하지 않다. 따라서 본 논문에서는 게임 플레이어가 제어하는 집단 게임캐릭터의 목표 지점까지의 길찾기를 위한 수정된 정렬 조타행동 알고리즘을 제안한다.

• The Journal of the Korea Contents Association
• /
• v.3 no.4
• /
• pp.11-16
• /
• 2003

In this paper, a Fast Block Sum Pyramid Algorithm (FBSPA) is presented for motion estimation in video coding. PBSPA is based on Block Sum Pyramid Algorithm(BSPA), Efficient Multilevel Successive Elimination Algorithms for Block Matching Motion Estimation, and Fast Algorithms for the Estimation of Motion Vectors. FBSPA reduces the computations for motion estimation of BSPA 29％ maximally using partial distortion elimination(PDE) scheme.

• Proceedings of the Korean Information Science Society Conference
• /
• 2005.07b
• /
• pp.694-696
• /
• 2005

본 논문에서는 개선된 유전자 알고리즘과 역전파 신경망 알고리즘의 특징을 살펴보고, 비선형 모의자료를 이용하여 개선된 유전자 알고리즘 기반의 신경망 학습 효과와 역전파 신경망 알고리즘을 이용한 신경망 학습 효과를 비교해 본다. 유전자 알고리즘을 이용한 신경망 학습에는 개선된 신경망 제어기를 이용한다. 역전파 알고리즘을 이용한 신경망 학습에는 일반화 성능향상을 위한 인자들의 결합효과를 이용한다. 모의실험을 통하여 두 가지의 학습에서 학습 수령의 정도와 학습 속도 등을 비교하는 모의실험 결과를 개선된 유전자 알고리즘과 신경망 알고리즘의 학습 결과와 항께 제시한다. 모의실험의 결과로서 유전자 알고리즘을 적용한 개선된 신경망 제어기를 통한 학습 결과가 일반 신경망 학습 결과보다 초기 가중값을 작은 범위에서 발생시킬 때 수렴 정확도 및 학습 속도에서 좋은 결과를 나타내 주고 있다.