• 한국시뮬레이션학회논문지
• /
• 제10권2호
• /
• pp.15-24
• /
• 2001

In this paper we study the delay patterns of voice traffic for end-to-end users Caused by serving the whole bunch of applications traffic at the same time on the Voice over Internet Protocol (VoIP) network. Given the current situation that voice traffic is served along with other application services on the VoIP network, it is quite necessary to figure out how and by what the voice traffic requiring high QoS is delayed. We compare the delay performance of voice traffic on the VoIP network under FIFO with the one under Weighted Fair Queuing(WFQ), and discover the differences of the delay performance resulting from the use of different voice codec algorithms. The results of our study show that using the voice codec algorithm with a higher coding rate nd the queuing algorithm of WEQ can provide users with high-quality voice traffic.

• 한국정보처리학회논문지
• /
• 제6권2호
• /
• pp.459-466
• /
• 1999

I broadband ISDN every packet will show different result if it would be processed according to its usage by the server. That is, normal data won't show big differences if they would be processed at normal speed. But it will improve the quality of service to process some kinds of data - for example real time video or voice type data or some data for a bid to by something through the internet - more fast than the normal type data. solution for this problem was suggested - priority packets. But the analyses of them are under way. Son in this paper a switching system for an output queuing model in a single server was assumed and some packets were given priorities and analysed. And correlation, simulating real life situation, was given too. These packets were analysed through three cases, first packets having no correlation, second packets having only correlation and finally packets having priority three cases, first packets having no correlation, second packets having only correlation and finally packets having priority and correlation. The result showed that correlation doesn't affect the mean delay time and the high priority packets have improved mean delay time regardless of the arrival rate. Those packets were assumed to be fixed-sized like ATM fixed-sized cell and the contention strategy was assumed to be short bus contention method for the output queue, and the mean delay length and the maximum 버퍼 length not to lose any packets were analysed.

• 정보처리학회논문지C
• /
• 제9C권1호
• /
• pp.21-30
• /
• 2002

오늘날 네트워크에서 다양한 애플리케이션의 서비스 성능을 저하시키는 불공정 큐잉 문제를 해결하기 위해 공정 큐잉 분야가 활발히 연구중이다. 그 중에서 DRR(Deficit Round Robin)은 작업 복잡도가 낮고 구현이 간단한 기법으로 매 라운드마다 각 큐에 대해 이전 라운드에서의 서비스 결손량을 포함한 서비스 할당량 SQ(Service Quantum)만큼을 서비스하도록 하여 다양한 트래픽들에 대해 정확한 공정성을 보장하는 기법이다. 그러나 엔터프라이스 네트워크 환경에서 최대 수 kbyte 이상의 패킷 사이즈를 가지는 서비스에 대해 불필요한 SQ 재설정 횟수 및 라운드 순회 횟수로 인한 지연시간 증가를 일으킨다. 본 논문에서는 이러한 DRR의 문제를 최소화하기 위해 SQ를 동적으로 설정하는 기법 및 패킷을 처리하는데 있어서의 작업 복잡도를 최소화하는 기법을 제안한다. 제안한 기법은 SQ의 동적 설정과 보다 단순한 패킷 처리로 인해 엔터프라이즈 네트워크 환경에서 다양한 애플리케이션들에 대한 지연시간을 최소화한다.

• ETRI Journal
• /
• 제19권4호
• /
• pp.344-362
• /
• 1997

The delay which is one of the quality of service parameters is considered to be a crucial factor for the effective usage of real-time audio and video streams in interactive multimedia collaborations. Among the various causes of the delay, we focus in this paper on the local delay concerned with the schemes which handle continuous inflow of encoded data from constant or variable bit-rate audio and video encoders. We introduce two kinds of implementation approaches, pull model and push model. While the pull model periodically pumps out the incoming data from the system buffer, the push model receives events from the device drivers. From our experiments based on Windows NT 3.51, it is shown that the push model outperforms the other for both constant and variable bit-rate streams in terms of the local delay, when the system suffers reasonable loads. We interpret this experimental data with M/G/1 multiple vacation queuing theories, and show that it is consistent with the queuing theoretic interpretations.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제10권10호
• /
• pp.4977-4996
• /
• 2016

The existence of excessively large and too filled network buffers, known as bufferbloat, has recently gained attention as a major performance problem for delay-sensitive applications. Researchers have made three types of suggestions to solve the bufferbloat problem. One is End to End (E2E) congestion control, second is deployment of Active Queue Management (AQM) techniques and third is the combination of above two. However, these solutions either seem impractical or could not obtain good bandwidth utilization. In this paper, we propose a Transmission Control Protocol（TCP）delayed window update mechanism which uses a congestion detection approach to predict the congestion level of networks. When detecting the network congestion is coming, a delayed window update control strategy is adopted to maintain good protocol performance. If the network is non-congested, the mechanism stops work and congestion window is updated based on the original protocol. The simulation experiments are conducted on both high bandwidth and long delay scenario and low bandwidth and short delay scenario. Experiment results show that TCP delayed window update mechanism can effectively improve the performance of the original protocol, decreasing packet losses and queuing delay while guaranteeing transmission efficiency of the whole network. In addition, it can perform good fairness and TCP friendliness.

• Journal of Communications and Networks
• /
• 제13권3호
• /
• pp.286-297
• /
• 2011

A novel converged optical network unit (ONU)-base station (BS) architecture has been contemplated for next-generation optical-wireless networks. It has been demonstrated through high quality studies that data traffic carried by both wired and wireless networks exhibit self-similar and long range dependent characteristics; attributes that classical teletraffic theory based on simplistic Poisson models fail to capture. Therefore, in order to apprehend the proposed converged architecture and to reinforce the provisioning of tightly bound quality of service (QoS) parameters to end-users, we substantiate the analysis of the QoS behavior of the ONU-BS under self-similar and long range dependent traffic conditions using custom queuing which is a common queuing discipline. This paper extends our previous work on priority queuing and brings novelty in terms of presenting performance analysis of the converged ONU-BS under realistic traffic load conditions. Further, the presented analysis can be used as a network planning and optimization tool to select the most robust and appropriate queuing discipline for the ONU-BS relevant to the QoS requirements of different applications.

• 한국통신학회논문지
• /
• 제37권6B호
• /
• pp.405-413
• /
• 2012

서비스 품질을 보장하기 위해 널리 사용되는 WFQ(Weighted Fair Queuing)는 엄격한 지연한계를 요구하는 저속 트래픽 흐름의 경우 자원 이용도가 아주 나빠지는 구조적인 한계를 갖고 있다. 이러한 WFQ의 문제점은 최적화 되지 않은 트래픽 흐름의 레이턴시에 기인하는 것으로 밝혀졌고, 최적 레이턴시를 실현하여 자원 이용도를 높일 수 있는 LOFQ(Latency-Optimized Fair Queuing) 가 도입되었다. 그러나 LOFQ는 새로운 흐름을 수락할 때마다 모든 흐름에 대해 최적의 레이턴시를 재 설정해야 하므로 흐름 수락시 O($N^2$)의 복잡도를 가져 구현에 어려움이 있다. 본 논문은 이러한 복잡도를 O(1) 수준으로 낮추는 방식을 제안한다. 제안된 방식은 가해진 트래픽 부하의 서비스 품질의 통계적인 특성을 이용하여 최적 레이턴시 지수 함수를 구한 후 이 함수를 이용하여 도착 흐름의 최적 레이턴시 지수를 간단하게 구하는 것이다.

• 한국컴퓨터정보학회논문지
• /
• 제16권11호
• /
• pp.45-56
• /
• 2011

WBAN은 사람이 착용하는 옷이나 인체 내부 혹은 외부에 있는 여러 장치들을 상호 연결하여 통신할 수 있는 근거리 무선통신 기술이다. WBAN은 기술 적용 분야에 따라 의료용과 비의료용으로 구분할 수 있으며, 주기적인 특성을 가진 의료용 데이터의 경우 GTS 방식을 사용해서 데이터를 전송하여 QoS를 보장한다. 본 논문에서는 WBAN 슈퍼프레임 구조에서 GTS 방식으로 데이터를 전송시 부족한 GTS 타임 슬롯을 해결하는 알고리즘을 제안한다. 코디네이터는 디바이스의 데이터 전송률에 따라 GTS 타임 슬롯을 가변적으로 할당하며, GTS의 개수가 부족하면 GTS 타임 슬롯을 공유하여 사용하도록 할당한다. 제안한 기법의 성능평가를 위해 지연한도와 처리량을 비교하였다. 즉, WBAN 환경에 적합한 WFQ(Weighted Fair Queuing) 스케줄링을 적용하여 상이한 데이터 전송률을 지원하면서 GTS를 할당하는 방법을 제안하였고, 라운드로빈(Round Robin) 스케줄링 방법과 비교하여 처리량은 증가하면서 최대 지연이 감소함을 알 수 있다.

• Journal of Communications and Networks
• /
• 제1권3호
• /
• pp.145-152
• /
• 1999

A new switch architecture is presented to enhance out-put queuing in photonic packet switches. Its appkication is for a packet switching enviroment based on the optical transport of fixed length packets. This architecture consists of a couple of cas-cading switching elements with output queuing, whose buffer ca-pacity is limited by photonic technology. The introduction of a suitable buffer management allows a very good and balanced ex-ploitation of the available optical memories, realized with fiber de-lay lines. In particular, packet loss performance is here evaluated showing the improvement with respect to the single switch and a way to design large optical switches is shown in order to meet broadband network requirements.

• 대한전자공학회:학술대회논문집
• /
• 대한전자공학회 2004년도 ICEIC The International Conference on Electronics Informations and Communications
• /
• pp.269-273
• /
• 2004

This paper presents a comparative performance analysis of a network accelerator model based on M/M/l queuing system. It assumes the Poisson distribution as its input traffic load. The decoding delay is employed as a performance analysis measure. Simulation results based on the proposed model show only $15\%$ differences with respect to actual measurements on field traffic for BCM5820 accelerator device. The performance analysis model provides with reasonable hardware structure of network servers, and can be used to span design spaces statistically.