• 대한수학회논문집
• /
• 제10권3호
• /
• pp.715-733
• /
• 1995

We present a transient queue size distribution for $M/G/m/N$ queue with state dependent arrival rates, using the diffusion process with piecewise constant diffusion parameters, with state space [0, N] and elementary return boundaries at x = 0 and x = N. The model considered here contains not only many basic model but the practical models such as as two-node cyclic queue, repairmen model and overload control in communication system with finite storage buffer. For the accuracy check, we compare the approximation results with the exact and simulation results.

• 한국정보처리학회논문지
• /
• 제6권11호
• /
• pp.3097-3107
• /
• 1999

Recently many studies have been done on the Random Early Detection(RED) algorithm as an active queue management and congestion avoidance scheme in the Internet. In this paper we first overview the characteristics of RED and the modified RED algorithms in order to understand the current status of these studies. Then we analyze the RED dynamics by investigating how RED parameters affect router queue behavior. We show the cases when RED fails since it cannot react to queue state changes aggressively due to the deterministic use of its parameters. Based on the RED parameter analysis, we propose a self-adaptive algorithm to cope with this RED weakness. In this algorithm we make two parameters be adjusted themselves depending on the queue states. One parameter is the maximum probability to drop or mark the packet at the congestion state. This parameter can be adjusted to react the long burst of traffic, consequently reducing the congestion disaster. The other parameter is the queue weight which is also adjusted aggressively in order for the average queue size to catch up with the current queue size when the queue moves from the congestion state to the stable state.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제13권3호
• /
• pp.1345-1362
• /
• 2019

Most active queue management algorithms manage network congestion based on the size of the queue but ignore the network environment which makes queue size change. It seriously affects the response speed of the algorithm. In this paper, a new AQM algorithm named CT-AQM (Change Trend-Adaptive Queue Management) is proposed. CT-AQM predicts the change trend of queue size in the soon future based on the change rate of queue size and the network environment, and optimizes its dropping function. Simulation results indicate that CT-AQM scheme has a significant improvement in loss-rate and throughput.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제9권8호
• /
• pp.2875-2893
• /
• 2015

Existing coding aware routing algorithms focused on novel routing metric design that captures the characteristics of network coding. However, in packet coding algorithm, they use opportunistic coding scheme which didn't consider the queue state of the coding node and are equivalent to the conventional store-and-forward method in light traffic load condition because they never delay packets and there are no packets in the output queue of coding node, which results in no coding opportunity. In addition, most of the existing algorithms assume that all flows participating in the network have equal rate. This is unrealistic since multi-rate environments are often appeared. To overcome above problem and expand network coding to light traffic load scenarios, we present an enhanced coding-aware routing algorithm based on queue state and local topology (OQMCAR), which consider the queue state of coding node in packet coding algorithm where the control policy is of threshold-type. OQMCAR is a unified framework to merge single rate case and multiple rate case, including the light traffic load scenarios. Simulations results show that our scheme can achieve higher throughput and lower end-to-end delay than the current mechanisms using COPE-type opportunistic coding policy in different cases.

• 한국통신학회논문지
• /
• 제24권9A호
• /
• pp.1354-1358
• /
• 1999

In this paper, we suggest a simple computational algorithm to obtain the queue length distribution in the finite queue, where the input process consists of several homogeneous two-state Markov modulated Poisson processes. With comparison to the conventional algorithm, is more practical, in particular, when a large number of input sources are loaded to the system.

• Journal of applied mathematics & informatics
• /
• 제22권1_2호
• /
• pp.453-460
• /
• 2006

A discrete-time priority queueing system with place reservation discipline is studied, in which two different types of packets arrive according to batch geometric streams. It is assumed that there is a reserved place in the queue. Whenever a high-priority packet enters the queue, it will seize the reserved place and make a new reservation at the end of the queue. Low-priority arrivals take place at the end of the queue in the usual way. Using the probability generating function method, the joint distribution of system state and the delay distribution for each type are obtained.

• Journal of the Korean Society for Industrial and Applied Mathematics
• /
• 제18권4호
• /
• pp.295-304
• /
• 2014

We analyze theMMPP/G/1/K queue with a modified state-dependent service rate. The service time of customers upon service initiation is changed if the number of customers in the system reaches a threshold. Then, the changed service time is continued until the system becomes empty completely, and this process is repeated. We analyze this system using an embedded Markov chain and a supplementary variable method, and present the queue length distributions at a customer's departure epochs and then at an arbitrary time.

• 대한산업공학회지
• /
• 제28권3호
• /
• pp.247-255
• /
• 2002

The objective of this paper is to clarify the decomposition property for $M^{X}$/G/1 queues with generalized vacations so that the decomposition property is better understood and becomes more applicable. As an example model, we use the $M^{X}$/G/1 queue with setup time. For this queue, we correct Choudhry's (2000) steady-state queue size PGF and derive the steady-state waiting time LST. We also present a meaningful interpretation for the decomposed steady-state waiting time LST.

• 한국통신학회논문지
• /
• 제28권9B호
• /
• pp.823-833
• /
• 2003

인터넷에서 실시간 서비스를 지원하기 위해서는 큐 관리 및 스케줄링 기법이 고도화되어야 한다. 기존의 큐 관리 기법은 TCP 플로우의 혼잡을 제어하는 기능은 제공하고 있으나 실시간 서비스 지원에는 문제점을 가지고 있다. 즉, 큐가 어느 정도 점유된 이후 계속적인 버스트 성의 트래픽이 발생할 경우 성능이 급격히 떨어지고, 특히 큰 지터 값으로 인해 실시간 전송에 부적절하다. 본 논문에서는 최대 임계값과 큐 가중치 등의 RED 파라미터를 큐 상태에 따라 적응적으로 조정하여 실시간 서비스에 대한 지연 및 지터 성능을 향상시킬 수 있는 상태 의존 RED (SDRED; State Dependent Random Early Detection) 큐 관리 및 동적 스케줄링 기법을 제안한다. 제안한 기법은 큐 상태에 따라 RED의 최대 임계값과 큐 가중치를 네 개의 서로 다른 레벨로 변경하여 트래픽 상황에 적응하도록 설계하였으며, 시뮬레이션을 통하여 SDRED가 평균 큐 크기를 가지는 RED나 평균 큐 크기를 가지지 않고 현재의 큐 크기를 사용하여 흔잡을 제어하는 BLUE, 그리고 RED 게이트웨이의 수율이나 지연 특성을 향상하기 위해 제안된 ARED 및 DSRED에 비해 낮은 지연 값과 안정적인 지터 값을 나타냄을 검증하였다.

• Journal of the Korean Statistical Society
• /
• 제31권4호
• /
• pp.491-507
• /
• 2002

We analyze a single server bulk input queue with optional server vacations under a single vacation policy and balking phenomenon. The service times of the customers as well as the vacation times of the server have been assumed to be arbitrary (general). We further assume that not all arriving batches join the system during server's vacation periods. The supplementary variable technique is employed to obtain time-dependent probability generating functions of the queue size as well as the system size in terms of their Laplace transforms. For the steady state, we obtain probability generating functions of the queue size as well as the system size, the expected number of customers and the expected waiting time of the customers in the queue as well as the system, all in explicit and closed forms. Some special cases are discussed and some known results have been derived.