• 제목/요약/키워드: random algorithm

검색결과 1,806건 처리시간 0.031초

라우터에서의 동적인 혼잡 제어를 위한 새로운 큐 관리 알고리즘 (Modified Random Early Defection Algorithm for the Dynamic Congestion Control in Routers)

  • 구자헌;송병훈;정광수;오승준
    • 한국정보과학회논문지:정보통신
    • /
    • 제28권4호
    • /
    • pp.517-526
    • /
    • 2001
  • 현재의 인터넷 라우터는 Drop tail 방법으로 패킷을 처리한다. 따라서 네트워크 트래픽의 지수적인 증가로 인한 혼잡상황 때문에 많은 패킷이 손실 될 수 있다. 이러한 문제를 해결하기 위해 IETF(Internet Engineering Task Force)에서는 RED(Random Early Detection)와 같은 능동적인 큐 관리 알고리즘을 제시하였다. 이러한 알고리즘은 라우터나 게이트웨이에서 발생할 수 있는 네트워크 혼잡상황을 개선하여 전체적인 패킷 손실을 줄여 줄 수 있다. 그러나 RED 알고리즘의 경우 큐 크기의 변화 정보로 혼잡상황을 판단하고 제어하기 때문에 동적으로 변화하는 현재의 인터넷 트래픽 대한 혼잡상황 제어에 있어서 개선의 필요성이 있다. 본 논문에서는 이러한 RED 알고리즘의 문제를 보완하기 위해서 기존의 RED 알고리즘을 개선한 MRED(Modified Random Early Detection) 알고리즘을 제안했다. 제안한 알고리즘은 간단하게 폐기(drop) 확률을 구하는 RED에 비하여 휴리스틱 (heuristic)한 방법을 적용하여 보다 동적으로 폐기 확률 값을 계산한다. MRED 알고리즘의 성능을 검증하기 위해서 실힘을 통하여 기존의 큐 관리 방법과 성능을 비교하였고 리눅스 커널에 MRED를 구현하여 성능을 분석하였다.

  • PDF

클러스터링과 마르코프 랜덤 필드를 이용한 배경 모델링 기법 제안 (Improving Clustering-Based Background Modeling Techniques Using Markov Random Fields)

  • 한희얼;박수빈
    • 대한전자공학회논문지SP
    • /
    • 제48권1호
    • /
    • pp.157-165
    • /
    • 2011
  • 본 논문에서는 마르코프 랜덤 필드(Markov random fields: MRF) 기반으로 배경을 모델링하는 방식과 함께 관련 파라미터들을 추정하는 알고리즘을 제안한다. 화소 기반의 배경 모델링 기법은 인근 화소 간의 연관성을 고려하지 않고 화소 단위의 시간적 변화에 대한 통계적 특성에 주로 의존하므로 판정 오류를 줄이는데 한계가 있다. 제안 알고리즘은 화소 기반으로 배경 모델을 일차적으로 수행한 다음 MRF를 이용하여 시공간적으로 인근한 화소 간의 상호 의존성을 활용하여 배경모텔의 정확도를 향상시키는데 그 목적을 두고 있다. MRF는 기본적으로 파라미터의 크기에 매우 민감하므로 기존의 MRF 기반 알고리즘은 이미지에 따라 적절한 값을 사전에 구하여 적용하고 있다. 제안한 방식은 초기에 임의의 파라미터로 배경/전경 상태변수를 구한 후에 이의 통계적 특성을 이용하여 파라미터들을 추정하고 추정된 파라미터를 적용하여 상대변수를 재차 구하는 과정을 반복함으로써 최적의 파라미터에 적응적으로 수렴하도록 조정한다. 실내외의 다양한 환경에서 촬영한 비디오를 이용하여 제안한 방식 성능을 확인한다.

HIPERLAN/2에서 최소 RCH 수를 제한한 동적 랜덤채널 할당 기법 (Dynamic Random Channel Allocation Algorithm by Limiting the Number of Minimum RCH in HIPERLAN/2)

  • 강재은;이정규
    • 한국통신학회논문지
    • /
    • 제34권1A호
    • /
    • pp.58-64
    • /
    • 2009
  • 본 논문에서는 최적화된 $\beta$개로 최소 랜덤 채널 RCH(Random CHannels) 개수를 제한함으로써, 단말이 채널 경쟁 구간에서 채널을 획득할 확률을 높이고 지연시간을 줄이는 알고리즘을 제안한다. HIPERLAN/2의 MAC 프로토콜은 TDMA/TDD 기반으로 동작하며, 단말은 RCH 구간을 통해 채널 경쟁에서 성공함으로써 트래픽 전송기회를 획득한다. 그러므로 AP는 각 RCH에서의 단말들의 채널 요청 메시지의 충돌 여부에 따라, 다음 프레임에서 최적의 RCH 개수를 동적으로 할당할 수 있다. 따라서 본 논문에서는 최소 RCH 개수를 $\beta$개로 제한하는 동적 랜덤채널할당 기법을 통해 채널 경쟁에서의 최소 성공률을 보장함으로써 HIPERLAN/2의 시스템 성능을 향상시킨다. 또한 실제 인터넷 트래픽에 기반한 이더넷 분포와 배치 도착을 적용한 시뮬레이션을 통해 최적의 $\beta$ 값을 찾아내며, 제안하는 알고리즘이 기존 알고리즘과 비교하여 처리율이 향상되고 최대 19%의 지연시간이 감소됨을 보여준다.

Correlation-based Feature Selection 기법과 Random Forest 알고리즘을 이용한 한강유역 지류의 TDI 예측 연구 (A Study on Predicting TDI(Trophic Diatom Index) in tributaries of Han river basin using Correlation-based Feature Selection technique and Random Forest algorithm)

  • 김민규;윤춘경;이한필;황순진;이상우
    • 한국물환경학회지
    • /
    • 제35권5호
    • /
    • pp.432-438
    • /
    • 2019
  • The purpose of this study is to predict Trophic Diatom Index (TDI) in tributaries of the Han River watershed using the random forest algorithm. The one year (2017) and supplied aquatic ecology health data were used. The data includes water quality(BOD, T-N, $NH_3-N$, T-P, $PO_4-P$, water temperature, DO, pH, conductivity, turbidity), hydraulic factors(water width, average water depth, average velocity of water), and TDI score. Seven factors including water temperature, BOD, T-N, $NH_3-N$, T-P, $PO_4-P$, and average water depth are selected by the Correlation Feature Selection. A TDI prediction model was generated by random forest using the seven factors. To evaluate this model, 2017 data set was used first. As a result of the evaluation, $R^2$, % Difference, NSE(Nash-Sutcliffe Efficiency), RMSE(Root Mean Square Error) and accuracy rate show that this model is compatible with predicting TDI. To be more concrete, $R^2$ is 0.93, % Difference is -0.37, NSE is 0.89, RMSE is 8.22 and accuracy rate is 70.4%. Also, additional evaluation using data set more than 17 times the measured point was performed. The results were similar when the 2017 data set were used. The Wilcoxon Signed Ranks Test shows there was no statistically significant difference between actual and predicted data for the 2017 data set. These results can specify the elements which probably affect aquatic ecology health. Also, these will provide direction relative to water quality management for a watershed that must be continuously preserved.

Nonlinear control system using universal learning network with random search method of variable search length

  • Shao, Ning;Hirasawa, Kotaro;Ohbayashi, Masanao;Togo, Kazuyuki
    • 제어로봇시스템학회:학술대회논문집
    • /
    • 제어로봇시스템학회 1996년도 Proceedings of the Korea Automatic Control Conference, 11th (KACC); Pohang, Korea; 24-26 Oct. 1996
    • /
    • pp.235-238
    • /
    • 1996
  • In this paper, a new optimization method which is a kind of random searching is presented. The proposed method is called RasVal which is an abbreviation of Random Search Method with Variable Seaxch Length and it can search for a global minimum based on the probability density functions of searching, which can be modified using informations on success or failure of the past searching in order to execute intensified and diversified searching. By applying the proposed method to a nonlinear crane control system which can be controlled by the Universal Learning Network with radial basis function(R.B.P.), it has been proved that RasVal is superior in performance to the commonly used back propagation learning algorithm.

  • PDF

Reliability sensitivities with fuzzy random uncertainties using genetic algorithm

  • Jafaria, Parinaz;Jahani, Ehsan
    • Structural Engineering and Mechanics
    • /
    • 제60권3호
    • /
    • pp.413-431
    • /
    • 2016
  • A sensitivity analysis estimates the effect of the change in the uncertain variable parameter on the probability of the structural failure. A novel fuzzy random reliability sensitivity measure of the failure probability is proposed to consider the effect of the epistemic and aleatory uncertainties. The uncertainties of the engineering variables are modeled as fuzzy random variables. Fuzzy quantities are treated using the ${\lambda}$-cut approach. In fact, the fuzzy variables are transformed into the interval variables using the ${\lambda}$-cut approach. Genetic approach considers different possible combinations within the search domain (${\lambda}$-cut) and calculates the parameter sensitivities for each of the combinations.

Numerical simulation of tuned liquid tank- structure systems through σ-transformation based fluid-structure coupled solver

  • Eswaran, M.;Reddy, G.R.
    • Wind and Structures
    • /
    • 제23권5호
    • /
    • pp.421-447
    • /
    • 2016
  • Wind-induced and earthquake-induced excitations on tall structures can be effectively controlled by Tuned Liquid Damper (TLD). This work presents a numerical simulation procedure to study the performance of tuned liquid tank- structure system through ${\sigma}$-transformation based fluid-structure coupled solver. For this, a 'C' based computational code is developed. Structural equations are coupled with fluid equations in order to achieve the transfer of sloshing forces to structure for damping. Structural equations are solved by fourth order Runge-Kutta method while fluid equations are solved using finite difference based sigma transformed algorithm. Code is validated with previously published results. The minimum displacement of structure is observed when the resonance condition of the coupled system is satisfied through proper tuning of TLD. Since real-time excitations are random in nature, the performance study of TLD under random excitation is also carried out in which the Bretschneider spectrum is used to generate the random input wave.

Bayesian analysis of random partition models with Laplace distribution

  • Kyung, Minjung
    • Communications for Statistical Applications and Methods
    • /
    • 제24권5호
    • /
    • pp.457-480
    • /
    • 2017
  • We develop a random partition procedure based on a Dirichlet process prior with Laplace distribution. Gibbs sampling of a Laplace mixture of linear mixed regressions with a Dirichlet process is implemented as a random partition model when the number of clusters is unknown. Our approach provides simultaneous partitioning and parameter estimation with the computation of classification probabilities, unlike its counterparts. A full Gibbs-sampling algorithm is developed for an efficient Markov chain Monte Carlo posterior computation. The proposed method is illustrated with simulated data and one real data of the energy efficiency of Tsanas and Xifara (Energy and Buildings, 49, 560-567, 2012).

Hyper-Parameter in Hidden Markov Random Field

  • Lim, Jo-Han;Yu, Dong-Hyeon;Pyu, Kyung-Suk
    • 응용통계연구
    • /
    • 제24권1호
    • /
    • pp.177-183
    • /
    • 2011
  • Hidden Markov random eld(HMRF) is one of the most common model for image segmentation which is an important preprocessing in many imaging devices. The HMRF has unknown hyper-parameters on Markov random field to be estimated in segmenting testing images. However, in practice, due to computational complexity, it is often assumed to be a fixed constant. In this paper, we numerically show that the segmentation results very depending on the fixed hyper-parameter, and, if the parameter is misspecified, they further depend on the choice of the class-labelling algorithm. In contrast, the HMRF with estimated hyper-parameter provides consistent segmentation results regardless of the choice of class labelling and the estimation method. Thus, we recommend practitioners estimate the hyper-parameter even though it is computationally complex.

2차원 불규칙 조면에서의 전자파 전파 해석 (Analysis of Electromagnetic Wave Propagation from 2 Dimensional Random Rough Surfaces)

  • 윤광렬
    • 전기학회논문지
    • /
    • 제59권6호
    • /
    • pp.1114-1119
    • /
    • 2010
  • This paper is concerned with an numerical analysis of electromagnetic wave propagation from randomly rough surfaces as a desert, sea surface and so on. We propose discrete ray tracing method (DRTM) for analysis of characteristics of wave propagation along one dimensional (1D) and two dimensional (2D) random rough surfaces. The point of the present method is to discretize not only rough surface but also ray tracing. This technique helps saving computer memories and does simplifying ray searching algorithm resulting in saving computation time. Numerical calculations are carried out for 1D and 2D random rough surfaces and electric field distributions are shown to check the effectiveness of the proposed DRTM.