• Proceedings of the Korea Society for Simulation Conference
• /
• 1998.10a
• /
• pp.82-85
• /
• 1998

본 논문에서는 음성호 및 데이터호를 동시에 서비스하는 계층구조를 가지는 CDMA 셀룰러 시스템에서 트래픽 특성을 고려한 효율적인 채널 할당 알고리듬을 제안한다. 제안된 채널할당 알고리듬은 첫째, 계층구조를 가지는 셀룰러 시스템에서 호의 종류를 음성호와 데이타호로 나누고 트래픽의 특성을 분석한다. 둘째, 위에서 구한 트래픽 특성을 바탕으로 핸드오프와 신규호 블록킹확률을 최소로 하기 위하여 마이크로셀에서 매크로셀로 오버플로우 시키는 방식을 적용한다. 셋째, 신규화와 핸드오프호의 블록킹 확률을 구하고 QoS를 만족시키는 음성호와 데이터호의 채널분배 방법을 제시하여 시스템의 성능을 향상시키고자 한다.

• Proceedings of the KIEE Conference
• /
• 2000.11d
• /
• pp.624-628
• /
• 2000

이 논문에서는 이산형 상태 공간 모델에서의 새로운 FIR 필터를 제안한다. 선형성, 무진동성, FIR 구조, 초기 조건과의 무관성등을 디자인 과정에서 고려해서 성능 지표를 최소화 하는 필터를 구한다. 성능지표로는 구간에서 외란 에너지와 현재 추정 에러의 최대 이득으로 생각하며, 이 지표는 일반적인 성능지표와는 다르다. 제안된 필터는 배치 형태로 먼저 구하고, 점화식 형태로 바꿀 수 있음을 보인다. 제안한 필터는 확률론적 시스템의 이동 구간 무편향 FIR 필터(RHUFF)와 유사함을 보인다.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2006.05a
• /
• pp.1835-1839
• /
• 2006

최근 빈번히 발생하는 국지성 집중호우 등의 기상이변과 도시화로 인한 불투수 면적의 증가는 급격한 유출량의 변화를 가져와 홍수로 인한 피해를 가중시킬 수 있다. 따라서 홍수방어시설인 제방의 경우 발생가능성은 낮지만 확률홍수량에 따른 제방 여유고의 평가도 필요하다. 본 연구에서는 대표적 도시하천인 탄천을 대상으로 하천정비기본계획(2000)에 수립된 계획홍수량과 최근 7년간 실측한 관측 결과로 계산한 확률홍수량을 조건으로 각각의 경우에 따른 한강 합류부(하류단)부터 상류 9 Km까지의 제방 여유고를 국내와 국외의 설계기준으로 비교, 검토하였다. 검토조건으로는 계획홍수량에 대해서는 3가지 경우의 하류단 수위를 경계조건으로 설정하였고 확률홍수량(80년, 100년, 200년 빈도)에 대해서는 각각의 경우에 발생 가능한 하류단 수위를 경계조건으로 설정하였다. 검토결과 탄천의 경우 계획홍수량에 대해서는 대체로 안전측으로 나타났지만 계획빈도 이상의 확률홍수량에 대해서는 국내, 일본의 설계기준에 제시된 여유고보다 상류 일부구간에서 여유고가 부족한 경향을 보이는 것으로 나타났다. 제방 여유고를 미국(도시하천의 경우 최소 0.91 m)의 경우와 비교하여도 안전측에 속한다고 평가할 수 있다. 하지만 본 검토의 대상인 제방이외의 일부 수공구조물(교량등)에서 여유고가 일부 부족한 것도 확인할 수 있었다. 이에 새로이 하천정비를 실시할 탄천(2006년 예정)의 경우 이러한 부분에 대한 검토 및 보완이 필요할 것으로 사료된다.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.10 no.3
• /
• pp.270-279
• /
• 2000

This paper describes the on-chip learning algorithm of neural networks using the stochastic pulse arithmetic. Stochastic pulse arithmetic is the computation using the numbers represented by the probability of 1' and 0's occurrences in a random pulse stream. This stochastic arithmetic has the merits when applied to neural network ; reduction of the area of the implemented hardware and getting a global solution escaping from local minima by virtue of the stochastic characteristics. And in this study, the on-chip learning algorithm is derived from the backpropagation algorithm for effective hardware implementation. We simulate the nonlinear separation problem of the some character patterns to verify the proposed learning algorithm. We also had good results after applying this algorithm to recognize printed and handwritten numbers.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.40 no.12
• /
• pp.2443-2449
• /
• 2015

Downlink (DL) data rate for a MS is influenced by not only the signal to interference and noise ratio (SINR) but also the amount of radio resources allocated to the MS. Therefore, when a MS uses SINR to select a cell to associate with, it cannot receive the fastest DL data rate all the time if it associates with a congested cell. Moreover, the SINR-based cell selection may result in cell loads unbalance, which decreases the efficiency of a network. To address the issue, we propose a novel cell selection method by considering not only SINR but also a cell load which are combined into two cell selection criteria. One is the maximum achievable data rate and the other is the minimum outage probability. The simulation results show that the cell selection based on the maximum achievable data rate is superior to the SINR-based method and the method using the minimum outage probability in terms of the system efficiency and the fairness in cell loads while the cell selection method based on the minimum outage probability is superior to the others in terms of the outage probability of a MS.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.20 no.3
• /
• pp.654-662
• /
• 2016

As the number of gates and basic events in fault trees increases, it becomes difficult to calculate the exact probability of the top event. In order to overcome this difficulty the BDD methodology can be used to calculate the exact top event probability for small and medium size fault trees in short time. Fault trees are converted to BDD by using CUDD library functions and a failure path search algorithm is proposed to calculate the exact top event probability. The backward search algorithm is more efficient than the forward one in finding failure paths and in the calculation of the top event probability. This backward search algorithm can reduce searching time in the identification of disjoint failure paths from BDD and can be considered as an effective tool to find the cut sets and the minimal cut sets for the given fault trees.

• The Journal of Korea Institute of Information, Electronics, and Communication Technology
• /
• v.9 no.3
• /
• pp.291-297
• /
• 2016

In this paper, we are comparative analysis of exit algorithm and proposal algorithm for desired target direction of arrival estimation in correlation signal system. Proposed algorithm in this paper is to decrease target direction of arrival an estimation error probability using bayesian, maximum posterior, and MUSIC algorithm in order to decrease direction of arrival error probability as optimize and use linear constrained minimum variance to update weight value. Through simulation, we were comparative analysis proposed algorithm and exit MUSIC algorithm. In case SNR is 10dB and antenna element arrays are 9 and 12, We show the superior performance of the proposed method relative to the class method to decrease of signal estimation error probability about 11% and 13%, respectively.

• The Journal of Korean Institute of Electromagnetic Engineering and Science
• /
• v.12 no.4
• /
• pp.570-575
• /
• 2001

The minimum duration outage probability of a cellular system is defined the probability that the received signal level is hold under minimum required threshold duration. This definition is more realistic compared to the outage probability which is the received signal strength level is below predefined threshold level. Especially, in a DS-CDMA cellular system the received signal-to-interference ratio is a function of cell traffic, while the former researches are only considered the minimum duration outage probability independent of the cell traffic. We noticed that the minimum outage probability is rapidly increases as cell traffic increase.

• The Journal of the Acoustical Society of Korea
• /
• v.15 no.4
• /
• pp.93-97
• /
• 1996

The Total Least Squares(TLS) method is an unbiased estimator for solving overdetermined sets of linear equations Ax${\simeq}$b when errors occur in all data. However, as well as Least Squares(LS) method it doesn't show robustness while the errors have a heavy tailed probability density function. In this paper we proposed a robust method of TLS (Robust TLS, ROTLS) based on the characteristics of TLS solution. And the ROTLS is verified by applying it to system identification problems.

• Journal of the Korean Data and Information Science Society
• /
• v.27 no.3
• /
• pp.711-724
• /
• 2016

Under variable complex operating environment, various factors can affect the lifetimes of systems. In this research, we study bivariate reliability models having multiple dynamic competing risks. As competing risks, in addition to the natural failure, we consider the increased stress caused by the failure of one component, external shocks, and the level of stress of the working environment at the same time. Considering two reliability models which take into account all of these competing risks, we derive bivariate life distributions. Furthermore, we compare these two models and also compare the distributions of maximum and minimum statistics in the two models.