• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.28 no.1A
• /
• pp.25-33
• /
• 2003

We derive the outage probabilities for the Erlang capacity and link margin based on handoffs in the reverse link of a CDMA system of 14.4 kbps/1.9 GHz band, calculate them with the same parameters, and then with the maximum realizable Erlang capacity analyze newly the cell coverage in different areas by the COST-231 model If it decreases from 5 % to 2 % in case of $E_b/N_0$=5 dB under a soft handoff, the result shows a very reliable link with additional increment of 0.88 dB in the margin and a high quality of service within the coverage decrease of 0.12 km, 0.25 km, and 1 km in a city, suburban, and open areas respectively on the decrease of 2 Erlang in the capacity. When the $E_b/N_0$ decreases from 7 dB to 5 dB within the outage probability of 5 %, it shows the increase of 9 Erlang and of 0.2 km in a city on the same margin of 2.8 dB. Then, comparing it with a single cell, it shows the decrease of more than 10 Erlang in the capacity, but the enlargement of 0.17 km, 0.3 km, and 1 km in them on the margin decrease of 1.31 dB, also, with a hard handoff, the soft handoff gain of the increase of 10 Erlang and of 0.22 km, 0.5 km, and 2 km on the decrease of 2.16 dB.

• The Korean Journal of Applied Statistics
• /
• v.11 no.2
• /
• pp.377-387
• /
• 1998

A Markov Chain Monte Carlo method with data augmentation is developed to compute the features of the posterior distribution. For each observed failure epoch, we introduced latent variables that indicates with component of the Superposition model. This data augmentation approach facilitates specification of the transitional measure in the Markov Chain. Metropolis algorithms along with Gibbs steps are proposed to preform the Bayesian inference of such models. for model determination, we explored the Pre-quential conditional predictive Ordinate(PCPO) criterion that selects the best model with the largest posterior likelihood among models using all possible subsets of the component intensity functions. To relax the monotonic intensity function assumptions, we consider in this paper Superposition of Musa-Okumoto and Erlang(2) models. A numerical example with simulated dataset is given.

• Journal of Information Technology Applications and Management
• /
• v.25 no.4
• /
• pp.115-122
• /
• 2018

Software reliability has the greatest impact on computer system reliability and software quality. For this software reliability analysis, In this study, we compare and analyze the trends of the properties affecting the reliability according to the shape parameters of Erlang distribution based on the finite fault NHPP. Software failure time data were used to analyze software failure phenomena, the maximum likelihood estimation method was used for parameter estimation. As a result, it can be seen that the intensity function is effective because it shows a tendency to decrease with time when the shape parameters a = 1 and a = 3. However, the pattern of the mean value function showed an underestimation pattern for the true values when the shape parameters a = 1 and a = 2, but it was found to be more efficient when a = 3 because the error width from the true value was small. Also, in the reliability evaluation of the future mission time, the stable and high trend was shown when the shape parameters a = 1 and a = 3, but on the contrary, when a = 2, the reliability decreased with the failure time. Through this study, the property of finite fault NHPP Erlang model according to the change of shape parameter without existing research case was newly analyzed, and new research information that software developers can use as basic guideline was presented.

• Journal of Digital Convergence
• /
• v.12 no.3
• /
• pp.219-226
• /
• 2014

In this study, reliability software cost model considering shape parameter based on life distribution from the process of software product testing was studied. The shape parameter using the Erlang and Log-logistic model that is widely used in the field of reliability problems presented. The software failure model was used finite failure non-homogeneous Poisson process model, the parameters estimation using maximum likelihood estimation was conducted. In comparison result of software cost model based on the Erlang distribution and the log-logistic distribution software cost model, because Erlang model is to predict the optimal release time can be software, but the log-logistic model to predict to optimal release time can not be, Erlang distribution than the log-logistic distribution appears to be effective. In this research, software developers to identify software development cost some extent be able to help is considered.

• International Journal of Reliability and Applications
• /
• v.1 no.1
• /
• pp.49-64
• /
• 2000

We consider a system whose inherent life follows an Erlang distribution, which is subject to two heterogeneous random shocks. Minor shocks arrive according to a renewal process and each causes the system to fail independently with a certain probability. A major shock whose interarrival times follow an Erlang distribution causes the system to fail with probability one. The Laplace transform of the distribution of the time to system failure is derived in a functional form of the Laplace transform of the interarrival time distribution of minor shocks. An algorithm is given for the computation of the moments of the time to system failure.

• Communications for Statistical Applications and Methods
• /
• v.12 no.2
• /
• pp.435-442
• /
• 2005

The mathematical properties of the sequentially operated systems are often described by integral equations. Reservoir system of a product and sequential probability ratio test (SPRT) are typical examples of sequentially operated systems. When the underlying random quantities follow Erlang distribution, a systematic method was developed to solve the integral equations. We extend the method to the cases having accrual functions of more general types. The solutions of the integral equations are represented as a linear combination of distribution functions, and the coefficients of the linear combination are obtained by solving linear system derived from the continuity condition of the solutions.

• Journal of the Korean Data and Information Science Society
• /
• v.20 no.1
• /
• pp.191-202
• /
• 2009

In this study, we drive the one dimensional marginal transform function, probability density function and probability distribution function for the random variables $T_{{\xi}N}$ (Time taken by the servers during the vacations), ${\xi}_N$(Number of vacations taken by the servers) and ${\eta}_N$(Number of customers or units arrive in the system) by controlling the variability of two random variables simultaneously.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.25 no.12A
• /
• pp.1806-1814
• /
• 2000

본 논문에서는 CDMA 계층구조 셀룰러 시스템에서 매크로 셀 기지국과 통신을 하는 고속 이동국과 마이크로 셀 기지국과 통신을 하는 저속 이동국의 비 그리고 소프트 핸드오프 영역 비에 따른 기지국이 받는 간섭전력과 시스템 용량을 구하였다. 매크로 셀 기지국과 통신을 하는 고속 이동국이 마이크로 셀 기지국과 통신을 하는 저속 이동국보다 많을수록 기지국이 받는 간섭전력은 증가하며 따라서 시스템 용량은 감소한다. 그러나 소프트 핸드오프 영역의 증가로 기지국이 받는 간섭전력은 감소하명 시스템 용량은 상당히 증가한다.

• Proceedings of the Korean Information Science Society Conference
• /
• 2004.10c
• /
• pp.64-66
• /
• 2004

콜 센터에서 CTI(Computer Telephony Integration)서버틀 통해 호(Call)를 제어할 때, 무한 입선 대기식 모델에서는 과잉호에 대한 지연확률이 발생한다. 이 지연 확률이 폴 센터에서 기준으로 삼은 시간 이후에는 호의 손실로 판단하게 되므로 호 접속 실패율을 최소로 하기 위한 노력이 필요하다. 본 논문에서는 AVAYA의 트래픽 계산기를 이용한 모의 실험을 통해 시간당 호 량과 대기시간 변화에 따른 호접속 실패를 최소로 하기 위한 trunk 수와 인원수에 대해 제시하였다.

• Proceedings of the Korean Information Science Society Conference
• /
• 2004.04b
• /
• pp.85-87
• /
• 2004

Erlang/OTP(Erlang Open Telecom Platform)의 Mnesia는 실시간 고장 허용 분산 MMDBMS이다. Mnesia는 메모리상에서의 질의 처리와 자동화된 데이타 분산 처리로 신속한 트랜잭션 처 리 가 가능하지 만 표준화된 DB 인터페이스와 SQL을 지원하지 않기 때문에 활용상에서의 한계를 갖고 있다. 본 논문에서는 SQL을 지원하는 java와 Mnesia의 연동에 대해 연구한다. 이를 위해 SQL컴파일러와 Mnesia인터페이스를 제공하는java라이브러리를 개발하였다. SQL를 이용한java와 Mnesia의 연동에 대 한 성능 분석 결과 성능의 우수성과 사용의 편리성을 모두 갖추게 되었다.