• Journal of Korea Society of Digital Industry and Information Management
• /
• v.6 no.2
• /
• pp.153-167
• /
• 2010

Traffic conditions, in particular number of active TCP flows, change dramatically over time. The main goal of this paper is an adaptive queue management algorithm that can maintain network state of high-throughput and low-delay under changing traffic conditions In this paper, we devise Probability Adaptive RED(PARED) that combines the more effective elements of recent algorithms with a RED core. It automatically adjusts its adaptive marking function to account for changes in traffic load and to keep queue length within the desired target queue length. We simulate that PARED algorithm results in under changes in traffic load and mixed traffic load. The simulation test confirm this stability, and indicate that overall performances of PARED are substantially better than the RED and ARED algorithms.

• Journal of Institute of Control, Robotics and Systems
• /
• v.12 no.6
• /
• pp.614-620
• /
• 2006

Today the automobile has been changed from a mechanical system to an electronic control system fly the development of the electronic technology. In the automobile body, most of these electronic control devices are networked and managed fully by the CAN protocol. But, when a network system is overloaded, unexpected transmission delay for relative low priority objects occurs due to the static priority definition of the CAN protocol. To resolve this problem, this paper proposes a dynamic precedence queue mechanism that creates a queue for the low priority object and its relevant objects to be transmitted, which becomes urgent in an overloaded network system to keep the maximum allowable time delay. For the generated queue, the highest priority is assigned to transmit the queued objects within the shortest time. The mechanism is implemented in the logical link layer of CAN, which does not require any modification of the old CAN hardware. Effectiveness of the proposed mechanism is verified by the real experiments with an automobile network system.

• Journal of the Korean Society for Precision Engineering
• /
• v.13 no.6
• /
• pp.152-160
• /
• 1996

This paper focuses on development and evaluation of a performance management algorithm for IEEE 802.4 token bus networks to serve large-scale integrated manufacturing systems. Such factory automation networks have to satisfy delay constraints imposed on time-critical messages while maintaining as much network capacity as possible for non-time-critical messages. This paper presents a network performance manager that adjusts queue capacity as well as timers by using a set of fuzzy rules and fuzzy inference mechanism. The efficacy of the performance management has been demonstrated by a series of simulation experiments.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2006.05a
• /
• pp.446-450
• /
• 2006

Recently, the advances on sensor technology increases the study on data processing oriented middleware. Usually, most of middleware uses the naive and delta method for queue management to process data effectively. But such a queue management method it is difficult to support guaranteeing of the requested quality of services because it simply discards the data in a queue when the overflow is occurred. To handle this problem, some methods for minimizing data volumes in a wireless sensor network have been studied, but most of them cause another problem that needs the huge processing time additionally with higher complexity of the proposed algorithm. In this paper, we propose a new method of delta-average queue management by applying the mean value to the exiting delta queue management method using the data differences to handle the problem of queue overflow. The proposed method has lower complexity than the others and increases the QoS of a WSN application by using mean value instead of using data discarding policy. In addition, it is designed to manage a queue effectively by controlling the range of data differences adaptively to the target sensor network environment.

• Journal of Communications and Networks
• /
• v.8 no.1
• /
• pp.93-105
• /
• 2006

The domination of the Internet by TCP-based services has spawned many efforts to provide high network utilization with low loss and delay in a simple and scalable manner. Active queue management (AQM) algorithms attempt to achieve these goals by regulating queues at bottleneck links to provide useful feedback to TCP sources. While many AQM algorithms have been proposed, most suffer from instability, require careful configuration of nonintuitive control parameters, or are not practical because of slow response to dynamic traffic changes. In this paper, we propose a new AQM algorithm, hybrid random early detection (HRED), that combines the more effective elements of recent algorithms with a random early detection (RED) core. HRED maps instantaneous queue length to a drop probability, automatically adjusting the slope and intercept of the mapping function to account for changes in traffic load and to keep queue length within the desired operating range. We demonstrate that straightforward selection of HRED parameters results in stable operation under steady load and rapid adaptation to changes in load. Simulation and implementation tests confirm this stability, and indicate that overall performances of HRED are substantially better than those of earlier AQM algorithms. Finally, HRED control parameters provide several intuitive approaches to trading between required memory, queue stability, and response time.

• Journal of KIISE:Information Networking
• /
• v.30 no.3
• /
• pp.437-447
• /
• 2003

In order to reduce the increasing packet loss rates caused by an exponential increase in network traffic, the IETF(Internet Engineering Task Force) is considering the deployment of active queue management techniques such as RED(Random Early Detection) algorithm. However, RED algorithm simple but does not protect traffic from high-bandwidth flows, which include not only flows that fail to use end-to-end congestion control such as UDP flow, but also short round-trip time TCP flows. In this paper, in order to solve this problem, we propose a simple fairness queue management scheme, called AFQM(Approximate Fair Queue Management) algorithm, that discriminate against the flows which submit more packets/sec than is allowed by their fair share. By doing this, the scheme aims to approximate the fair queueing policy Since it is a small overhead and easy to implement, AFQM algorithm controls unresponsive or misbehaving flows with a minimum overhead.

• ETRI Journal
• /
• v.25 no.1
• /
• pp.34-40
• /
• 2003

This paper considers a TMN-based management system for the management of public ATM switching networks using a four-level hierarchical structure consisting of one network management system, several element management systems, and several agent-ATM switch pairs. Using Jackson's queuing model, we analyze the effects of one TMN command on the performance of the component ATM switch in processing local calls. The TMN command considered is the permanent virtual call connection. We analyze four performance measures of ATM switches- utilization, mean queue length and mean waiting time for the processor directly interfacing with the subscriber lines and trunks, and the call setup delay of the ATM switch- and compare the results with those from Jackson's queuing model.

• Journal of Internet Computing and Services
• /
• v.2 no.2
• /
• pp.65-70
• /
• 2001

In this paper, a Modified RED algorithm for congestion avoidance in IP networks is presented. The RED detects incipient congestion by computing the average queue size. By notifying only a randomly selected fraction of connection, it causes to the global synchronization or fairness problem, And also, the network characteristics need to be known in order to find th optimum average queue length. When the average queue size exceeds a minimum threshold, a modified RED algorithm drops packets based on the state of each connection. Performance is improved because of keeping the average queue size low while allowing occasional bursts of packets in the queue, we compare performance of modified RED with RED and Drop Tail in terms of goodput, network utilization and fairness.

• Journal of the Korean Operations Research and Management Science Society
• /
• v.19 no.3
• /
• pp.1-14
• /
• 1994

In this paper, an analytical model of the message delay in the DQDB (Distributed Queue Dual Bus) network is investigated. The DQDB network has been adopted as a subnetwork for the IEEE 802 MAN (Metropolitan Area Network) standard. The DQDB network consists of two high speed undirectional buses and a series of stations attached to both of the buses. Massages arriving at each station consists of severla packets according to its size. This system is approximated into " $B^{[x]}$/G/1 with exceptional first service queueing " by defining the concept of service time on a packet. The service time for a packet is defined as the time from the instant the packet arrives at the transmission buffer until the time the packet is fully transmitted. By using the BASTA property and the average work in the system, the mean message delay time is obtained.age work in the system, the mean message delay time is obtained.d.

• Journal of Korea Multimedia Society
• /
• v.18 no.11
• /
• pp.1375-1382
• /
• 2015

This paper aims to introduce the numerical methods for solving the optimal control theory to model bufferbloat problem. Mathematical tools are useful to provide insight for system engineers and users to understand better about what we are facing right now while experiment in a large-scale testbed can encourage us to implement in realistic scenario. In this paper, we introduce a survey of the numerical methods for solving the optimal control problem. We propose the dynamic optimization sweeping algorithm for optimal control of the active queue management. Simulation results in network simulator ns2 demonstrate that our proposed algorithm can obtain the stability faster than the others while still maintain a short queue length (≈10 packets) and low delay experience for arriving packets (0.4 seconds).