• 전자공학회논문지 IE
• /
• v.43 no.3
• /
• pp.41-48
• /
• 2006

In this paper, we propose a new active queue management (AQM) scheme by utilizing the predictability of the Internet traffic. The proposed scheme predicts future traffic input rate by using the auto-regressive (AR) time series model and determines the future congestion level by comparing the predicted input rate with the service rate. If the congestion is expected, the packet drop probability is dynamically adjusted to avoid the anticipated congestion level. Unlike the previous AQM schemes which use the queue length variation as the congestion measure, the proposed scheme uses the variation of the traffic input rate as the congestion measure. By predicting the network congestion level, the proposed scheme can adapt more rapidly to the changing network condition and stabilize the average queue length and its variation even if the traffic input level varies widely. Through ns-2 simulation study in varying network environments, we compare the performance among RED, Adaptive RED (ARED), REM, Predicted AQM (PAQM) and the proposed scheme in terms of average queue length and packet drop rate, and show that the proposed scheme is more adaptive to the varying network conditions and has shorter response time.

• Journal of Advanced Navigation Technology
• /
• v.18 no.5
• /
• pp.461-467
• /
• 2014

Active queue management (AQM) algorithms run on routers and detect incipient congestion by typically monitoring the instantaneous or average queue size. When the average queue size exceeds a certain threshold, AQM algorithms infer congestion on the link and notify the end systems to back off by proactively dropping some of the packets arriving at a router or marking the packets to reduce transmission rate at the sender. Among the existing AQM algorithms, random early detection (RED) is well known as the representative queue-based management scheme by randomizing packet dropping. To reduce the number of timeouts in TCP and queuing delay, maintain high link utilization, and remove bursty traffic biases, the RED considers an average queue size as a degree of congestions. However, RED do not well in the specified networks conditions due to the fixed parameters($P_{max}$ and $TH_{min}$) of RED. This paper addresses a extended RED to be adapted in various networks conditions. By sensing network state, $P_{max}$ and $TH_{min}$ can be automatically changed to proper value and then RED do well in various networks conditions.

• The Journal of the Korea institute of electronic communication sciences
• /
• v.15 no.6
• /
• pp.1099-1104
• /
• 2020

In order to provide various types of application services based on the Internet, it is ideal to guarantee the quality of service(QoS) for each flow. However, realizing these ideas is not an easy task.. It is effective to classify multiple flows having the same or similar service quality requirements into same group, and to provide service quality for each group. The queue management mechanism in the router plays a very important role in order to efficiently transmit data and to support differentiated quality of service for each service. In order to efficiently support various multimedia services, an intelligent and adaptive queue management mechanism is required. This paper proposes an intelligent queue management mechanism based on deep reinforcement learning that decides whether to deliver packets for each group based on the traffic information of each flow group flowing in for a certain period of time and the current network state information.

• Proceedings of the Korean Operations and Management Science Society Conference
• /
• 1987.10a
• /
• pp.35-35
• /
• 1987

Consider a job shop that is modelled as an open queueing network of the Jackson(l957) type. All work stations in the shop have the same number of parallel servers. Two problems are studied : the loading of stations and the assignment of servers, which are represented by loading and assingment vectors, respectively. Ma jorization and arrangement orderings are established to order, respectively, the loading and the assignment vectors. It is shown that reducing the loading vector under ma jorizat ion or increasing the assignment vector under arrangement ordering will reduce the congestion in the shop in terms of reducing the total number of jobs(in the sense of likelihood ratio ordering), the maximum queue length(in the sense of stochastic ordering), and the queue-length vector( in the sense of stochastic majorization). The results can be used to supprot production planning in certain job shops, and to aid the desing of storage capacity. (OPEN QUEUEING NETWORK; WJORIZATION; ARRANGEMENT ORDERINC; LIKELIHOOD RATIO ORDERINC; STOCHASTIC ORDERING)

• Proceedings of the KIEE Conference
• /
• 2007.07a
• /
• pp.1668-1669
• /
• 2007

In this paper, we propose nonlinear congestion controller for active queue management of differentiated-services networks. Two important issues in differentiated-services architecture are bandwidth guarantee and fair sharing of unsubscribed bandwidth among TCP flows with and without bandwidth reservation. The nonlinear congestion controller was composed fuzzy logic controller and state feedback controller. The nonlinear congestion controller methodology has been applied to a TCP network. We use NS-2 simulation to demonstrate that the proposed control methodology achieves the desired behavior of the network, and possesses important attributes.

• Proceedings of the IEEK Conference
• /
• 2002.07a
• /
• pp.129-132
• /
• 2002

To support UDP-based real-time multimedia applications over the Internet, it is necessary to provide a certain amount of bandwidth within the network so that the performance of the applications will not be seriously affected during periods of congestion. Since the flow rates of some of these applications do not back of during periods of congestion, it is also necessary to protect flow-controlled TCP flows from unresponsive or aggressive UDP flows. To achieve these goals, we propose a simple queue policy to support multimedia applications, called threshold-based queue management (TBQM). TBQM isolates UDP flows efficiently from TCP flows to protect TCP flows while supporting bandwidth requirements of UDP applications that require QoS. In addition, TBQM supports drop fairness between TCP flows without maintaining per-flow state. We also present some experimental results to show that the proposed queue policy can work well.

• Journal of Korea Multimedia Society
• /
• v.18 no.12
• /
• pp.1475-1482
• /
• 2015

In network routers, buffers are used to resolve congestion and reduce packet loss rate whenever congestion occurs at bottleneck link. Most of the existing methods to manage such buffers focus only on queue-length-based control as one loop which have some issues of low link utilization and system stability. In this paper, we propose a novel framework which exploits two-loop control method, e.g. queue-length and congestion window size, combined with optimal margin method to facilitate parameter choices. Simulation results in ns-2 demonstrate that bottleneck link performance can be improved with higher link utilization (85%) and shorter queue length (22%) than the current deployed scheme in commercial routers (RED and DropTail).

• Proceedings of the KIEE Conference
• /
• 2006.07d
• /
• pp.1807-1808
• /
• 2006

In this paper, we present the wavelet neural network (WNN) controller as an active queue management(AQM) in end-to-end TCP network. AQM is important to regulate the queue length and short round trip time by passing or dropping the packets at the intermediate routers. As the role of AQM, the WNN controller adaptively controls the dropping probability of the TCP network and is trained by gradient-descent algorithm. We illustrate our result that WNN controller is superior to PI controller via simulations.

• Journal of KIISE:Information Networking
• /
• v.36 no.5
• /
• pp.425-436
• /
• 2009

In recent years, the user demands have increased for multimedia service of high quality over the broadband convergence network. These rising demands for high quality multimedia service led the popularization of various user terminals and large scale display equipments, which needs a variety type of QoS (Quality of Service). In order to support demands for QoS, numerous research projects are in progress both from the perspective of network as well as end system; For example, at the network perspective, QoS guaranteeing by improving of internet performance such as Active Queue Management, while at the end system perspective, SVC (Scalable Video Coding) encoding scheme to guarantee media quality. However, existing AQM algorithms have problems which do not guarantee QoS, because they did not consider the essential characteristics of video encoding schemes. In this paper, it is proposed to solve this problem by deploying the TS- AQM (Temporal Scalability Active Queue Management) which employs the differentiated packet dropping for dependency of the temporal level among the frames, based on SVC encoding characteristics by exploiting the TID (Temporal ID) field of the SVC NAL unit header. The proposed TS-AQM guarantees multimedia service quality through video decoding reliability for SVC streaming service, by differentiated packet dropping when congestion exists.

• 제어로봇시스템학회:학술대회논문집
• /
• 2004.08a
• /
• pp.782-787
• /
• 2004

Internet Engineering Task Force (IETF) has been considering the deployment of the Random Early Detection (RED) in order to avoid the increasing of packet loss rates which caused by an exponential increase in network traffic and buffer overflow. Although RED mechanism can prevent buffer overflow and hence reduce an average values of packet loss rates, but this technique is ineffective in preventing the consecutive drop in the high traffic condition. Moreover, it increases a probability and average number of consecutive dropped packet in the low traffic condition (named as "uncritical condition"). RED mechanism effects to TCP congestion control that build up the consecutive of the unnecessary transmission rate reducing; lead to low utilization on the link and consequently degrade the network performance. To overcome these problems, we have proposed a new mechanism, named as Extended Drop slope RED (ExRED) mechanism, by modifying the traditional RED. The numerical and simulation results show that our proposed mechanism reduces a drop probability in the uncritical condition.