Active Queue Management Algorithm Research Articles

A simple active queue management based on the prediction of the packet arrival rate

In this paper, we predict the change in the packet arrival rate at the link through the analysis of the network congestion control mechanism. An appropriate expression for dropping probability is derived based on this prediction to stabilize the queue length to the desired value. A new Active Queue Management (AQM) algorithm named as Straightforward AQM (SFAQM) is proposed with simple implementation. Its analysis of the stability is also carried out, and the necessary and sufficient condition for the system to be stable is presented. The simulation results using NS2 demonstrate that SFAQM is effective in different network environments. Compared to the conventional AQM controllers, such as PI, RaQ and CARED, SFAQM has smaller queue length jitter and faster convergence speed.

A resilient buffer allocation scheme in active queue management: a stochastic cooperative game theoretic approach

SummaryDuring the last decade, a plentiful number of active queue management schemes have been proposed, but their main objectives are simply allocating the buffer resource to all flows evenly, or protecting responsive flows from being degraded by unresponsive flows. However, the sending rates of the responsive flows can be determined diversely, and not all unresponsive flows have aggressively high sending rates. Furthermore, it is rational to reserve a portion of the buffer resource for certain privileged traffic. Grounded by these evidences, in this paper, we present a resilient active queue management algorithm, named Prior‐Core‐based Buffer Allocation considering diverse congestion control algorithms, fair‐unresponsive flows, and some privileged traffic. Our approach is based on stochastic cooperative game theory, where the payoffs yielded by cooperation are described by random variables, and the core is defined only over the distribution of these random payoffs; the core in this situation is called the prior‐core. As a result, it is shown that our buffer allocation, yielded by the prior‐core, achieves completely fair allocation for those flows whose requirement does not exceed the fair‐share regardless of the responsiveness, whereas aggressive flows are restricted according to availability of the buffer; all these are verified through ns‐2 simulation experiments. Copyright © 2014 John Wiley & Sons, Ltd.

CHOKeR: A Novel AQM Algorithm With Proportional Bandwidth Allocation and TCP Protection

Although differentiated services (DiffServ) networks have been well discussed in the past several years, a conventional Active Queue Management (AQM) algorithm still cannot provide low-complexity and cost-effective differentiated bandwidth allocation in DiffServ. In this paper, a novel AQM scheme called CHOKeR is designed to protect TCP flows effectively. We adopt a method from CHOKeW to draw multiple packets randomly from the output buffer. CHOKeR enhances the drawing factor by using a multistep increase and single-step decrease (MISD) mechanism. In order to explain the features of CHOKeR, an analytical model is used, followed by extensive simulations to evaluate the performance of CHOKeR. The analytical model and simulation results demonstrate that CHOKeR achieves proportional bandwidth allocation between different priority levels, fairness guarantee among equal priority flows, and protection of TCP against high-speed unresponsive flows when network congestion occurs.

Input-Rate Based Adaptive Fuzzy Neuron PID Control for AQM

Internet routers play an important role during network congestion. All the routers have buffers at input and output ports to hold the packets at congestion. Various congestion control algorithms have been proposed to control the congestion. Recently, some proportional-integral-derivative (PID) controller based algorithms have been proposed as Active Queue Management (AQM) schemes to address performance degradations of end-to-end TCP congestion control. However, most of the proposed PID-controllers for AQM are validated for their performance and stability via intuitive explanation and simulation studies instead of theoretic analysis and performance evaluation. But there are a few drawbacks of PID-controller based AQM algorithms leading to poor performance like causing data retention dropping and oscillation when the time delay is large, which means that the existing PID-controller can not meet the Quality of Service (QoS) requirements. To overcome the drawbacks of traditional PID, we analyze and enhance the PID-controller based AQM algorithm by regarding the TCP congestion control mechanism as an input-rate based Adaptive Fuzzy Neuron PID control algorithm (IRAFNPID) to avoid congestion in TCP/AQM networks. By means of simulations, we evaluate and compare the performance of traditional PID, single neural adaptive PID(SNAPID) and IRAFNPID, simulations with experiment data analysis and find that IRAFNPID has better convergence, stability, robustness, goodput and lower loss ratio.

CONGESTION CONTROL IN COMPUTER NETWORKS USING A MODIFIED RED AQM ALGORITHM

Congestion control plays an important role in netwo rk resource management in very large networks with heavy traffic. Active Queue Management (AQM) algorithms are one of the approach es to resolve the congestion problems. Majority of AQM algorithms mainly focus on single queued links. The input queued swit ches are limited in throughput and output queued sw itches require a large speedup factor so our attention is directed towards Combine d Input and Output Queued (CIOQ) switches. A simple modification to the RED AQM algorithm is proposed in this paper in order to account for the presence of both input and output queues in the switch. The weighted sum of input and output queue lengths are specifically used as the congestion measure instead of just the output queue length. The average backlog in the switch is signif icantly reduced in the low speedup region by using our modified algorithm as compared to RED without this modification. Simulati ons show that the loss rate in the modified RED sli ghtly larger than that in traditional RED but the output queue length in modi fied RED is tremendously reduced. The congestion me asure which is computed using weighting factor results in reduction of the average backlog. Using our modified algorithm simul ations indicate improvement in the queue length and switch utilization.

Active queue management algorithm based on particle swarm optimization

In order to mitigate the network congestion,a new active queue management algorithm named RQQM(Rate and Queue-based Queue Management) was proposed by Particle Swarm Optimization(PSO).In this algorithm,the actual queue length was deducted with PSO and variation factor,and the dropping strategy and dropping rate were given based on arrival rate and actual queue length.Then,a simulation with actual data was conducted to compare the algorithm performance between RQQM algorithm and RFED(Rate-based Fair Early Detection) algorithm,as well as ABLUE(Adaptive BLUE) algorithm.The results show that the dropping rate is greatly influenced by the utilization rate and buffer size,and the fairness of RQQM is much better than that of the other two algorithms,its average packet loss rate is decreased to 12.21%.

AQM controller design for TCP networks based on a new control strategy

When the network suffers from congestion, the core or edge routers signal the incidence of congestion through the active queue management (AQM) to the sources. The time-varying nature of the network dynamics and the complex process of retuning the current AQM algorithms for different operating points necessitate the development of a new AQM algorithm. Since the non-minimum phase characteristics of the network dynamics restrict direct application of the proportional-integral-derivative (PID) controller, we propose a compensated PID controller based on a new control strategy addressing the phase-lag and restrictions caused by the delay. Based on the unstable internal dynamics caused by the non-minimum phase characteristics, a dynamic compensator is designed and a PID controller is then allowed to meet the desired performance objectives by specifying appropriate dynamics for the tracking error. Since the controller gains are obtained directly from the dynamic model, the designed controller does not require to be tuned over the system operating envelop. Moreover, simulation results using ns2 show improvements over previous works especially when the range of variation of delay and model parameters are drastic. Simplicity, low computational cost, self-tuning structure and yet considerable improvement in performance are exclusive features of the proposed AQM for the edge or core routers.

RED-FT: A Scalable Random Early Detection Scheme with Flow Trust against DoS Attacks

In traditional Active Queue Management algorithms, e.g. RED, each flow, defined with the source and destination IP address of packets, fairly contends for the cache of bottleneck queues. However a malicious flow cannot be identified. And it enables potential network-layer attacks, e.g. the flooding Denial-of-Service (DoS) attack and the low-rate DoS attack. In this letter, we propose a new scheme using the flows trust values to defend against DoS attacks. Different from previous schemes, it employs the flow trust to safeguard legitimate flows. A router monitors network flows and calculates flows trust values, which are used for the relevant queue management. Malicious flows would be with lower trust values while legitimate flows would be with higher ones. Simulation results show that the scheme improves the throughput and delay in DoS attacking scenarios comparing with existing queue management algorithms. We consider the scheme is practical and effective to secure networks.

Adaptive mechanism-based congestion control for networked systems

In order to assure the communication quality in network systems with heavy traffic and limited bandwidth, a new ATRED (adaptive thresholds random early detection) congestion control algorithm is proposed for the congestion avoidance and resource management of network systems. Different to the traditional AQM (active queue management) algorithms, the control parameters of ATRED are not configured statically, but dynamically adjusted by the adaptive mechanism. By integrating with the adaptive strategy, ATRED alleviates the tuning difficulty of RED (random early detection) and shows a better control on the queue management, and achieve a more robust performance than RED under varying network conditions. Furthermore, a dynamic transmission control protocol–AQM control system using ATRED controller is introduced for the systematic analysis. It is proved that the stability of the network system can be guaranteed when the adaptive mechanism is finely designed. Simulation studies show the proposed ATRED algorithm achieves a good performance in varying network environments, which is superior to the RED and Gentle-RED algorithm, and providing more reliable service under varying network conditions.

SGNFL: An Active Queue Management Algorithm Based on Fuzzy-Logic for Network Communications in Smart Grid

SGNFL: An Active Queue Management Algorithm Based on Fuzzy-Logic for Network Communications in Smart Grid

On the use of a full information feedback to stabilize RED

RED and most of its family algorithms use only the average queue length as a congestion meter. Since the average queue length considers only long-term behavior of the queue, these algorithms fail to see instantaneous changes of the queue length and hence their reaction to the congestion is not fast enough. In other words the feedback generated by using only the average queue length does not reflect the network congestion precisely and hence leads to a poor performance and stability. This paper solves this problem by designing a RED-based active queue management (AQM) algorithm, called FUF-RED that provides a Full Information Feedback. This algorithm not only considers the average queue length but also it takes into account growth rate of the instantaneous queue length to calculate its congestion feedback. The proposed algorithm is supported by a theoretical stability analysis which gives those feedback gains that guarantees the network stability. Extensive packet level simulations, done by using ns-2 simulator, show that the proposed algorithm outperforms existing AQM algorithms in terms of stability, average queue length, number of dropped packets and bottleneck utilization.

A simulation study of a new rate‐and‐queue‐based active queue management algorithm

PurposeNowadays, e‐queues are built up everywhere where customer online service is necessary such as in banks' e‐service, enterprises' e‐business, etc. In order to enhance quality of service (QoS), active queue management (AQM) algorithms are frequently employed due to their efficiency in congestion avoidance as well as the differentiated forwarding of packets. This paper aims at developing a novel AQM algorithm to better QoS in terms of congestion prediction, queuing delay, packet loss and link utility, etc.Design/methodology/approachUpon the traditional designs of AQM, this paper establishes a new integrated AQM scheme (RQ‐AQM) by employing input rate and current queue length to calculate the packet dropping/marking probability. In this way, the rate feedback control enables to rapid response to congestion, decreasing the packet loss from buffer overflow. Meanwhile, the queue length feedback control stabilizes the queue length around a given target, achieving predictable queuing delay and lower delay jitter. Thus, the main feature of the design is to use coefficients of both proportional rate control and proportional‐integral queue length control, and to simplify parameter setting, the control parameters were scaled by the link capacity C to normalize the rate and by the bandwidth‐delay product BDP to normalize the queue length, respectively.FindingsThe stability performance of RQ‐AQM was tested via simulation under several conditions. The results proved that it is able to maintain the queue length around the given target. Also, the comparison results with other AQM schemes, including RED, ARED, PI controller, AVQ and REM, demonstrated the superiority of RQ‐AQM in low packet loss, faster convergence to target queue length and closest to the target queue length.Research limitations/implicationsThe main limitation of this study is that all the simulations were merely under a single bottleneck network topology. Furthermore, the system stability was examined under just a few cases. Other cases like TCP connections mixed with HTTP connections, or UDP flows, etc. can also be tested. Furthermore, the multiple bottleneck scenarios should be covered in the future work with more parameters set to enhance the proved results.Practical implicationsThe paper sets clear but ideal conditions for the performance of proposed algorithm; so the simulation results can only be used as a rough reference instead of an exact practical one. But the concepts the paper attempted to advocate could be considered seriously.Social implicationsThe scope of the paper is within the general theory of AQM. So it can be referred to any specific field that employs AQM technology, no matter locally or globally.Originality/valueThere are not much new brand contents in the paper. The main contribution is on some extension of the known related work.

Extended Queue Management Backward Congestion Control Algorithms for Web Transfers in Wireless Environment

Wireless links are characterized by high error rates and intermittent connectivity. TCP congestion control has been developed on the assumption that network congestion is the only cause for packet loss. Upon detecting a packet loss, TCP drops its transmit window resulting in an unnecessary reduction of end-to-end throughput which results in suboptimal performance.The sender has to be made aware by some feedback mechanism that some of the losses reported are not due to congestion. The Active Queue Management algorithms (AQM) are used to reduce congestion, and in this paper, we have analysed four AQM algorithms, Random Early Deduction (RED), Wireless Explicit Congestion Notification (WECN), Queue Management Backward Congestion Control Algorithm (QMBCCA) and its enhanced version Extended Queue Management Backward Congestion Control Algorithm (EQMBCCA). WECN, QMBCCA & EQMBCCA algorithms make use of feedback mechanisms. WECN gives feedback using the CE bit. QMBCCA and EQMBCCA make use of ISQ notifications and also the CE bit whenever the average queue size crosses minimum threshold value. EQMBCCA reduces the reverse ISQ traffic by introducing a configurable intermediate threshold value IntThres. The comparison is made in terms of Delay, HTTP packet loss percentage and fairness for FTP flows in a wireless environment. It is found that the performance of EQMBCCA is almost equal to that of QMBCCA and better than RED and WECN.

AQM Algorithm with Adaptive Reference Queue Threshold for Communication Networks

Nowadays, congestion in communication networks has been more intractable than ever before due to the explosive growth of network scale and multimedia traffic. Active queue management (AQM) algorithms had been proposed to alleviate congestion to improve quality of service (QoS), but existing algorithms often suffer from some flaws in one aspect or another. In this paper, a novel AQM algorithm with adaptive reference queue threshold (ARTAQM) is proposed of which the main innovative contributions are recounted as follows. First, traffic is predicted to calculate the packet loss ratio (PLR) and the traffic rate based on traffic prediction algorithm. Second, by means of periodical measurements, a weighted PLR is obtained to dynamically adjust packet dropping probability in ARTAQM algorithm. Third, ARTAQM algorithm runs in both coarse and fine granularities. In coarse granularity, the mismatch of the predicted traffic rate and link capacity can adjusts the reference queue length in every period, while in fine granularity, reference queue remains fixed and the instantaneous queue is adjusted packet by packet in one period. Simulation results indicate that ARTAQM algorithm not only maintains stable queue and fast response speed, but has lower PLR and higher link utilization as well. DOI: http://dx.doi.org/10.11591/telkomnika.v10i5.1274

IAPI: An intelligent adaptive PI active queue management scheme

Internet traffic is bursty and unpredictable. It is therefore important to control congestion to protect users from traffic surges in a way that is efficient, stable and robust. Research has shown that active queue management (AQM) has the potential to control congestion such that these requirements are met. However, despite much work and many published proposals of AQM algorithms, a satisfactory solution is still unavailable. In this paper, an intelligent adaptive PI controller (IAPI) is proposed as a new AQM scheme, and its design details are presented. Setting of parameter values for IAPI is provided based on previous studies on PI and additional empirical studies. The performance of IAPI is evaluated by simulations and the results demonstrate that it is stable and robust under various scenarios including cases involving different types of background traffic (UDP and HTTP) and a case involving a multiple bottlenecks. Its superiority over other AQMs is also demonstrated.

Evolution and Optimization of Active Queue Management Algorithms over High Bandwidth Aggregates

ABSTRACT Due to enormous growth of the Internet, demands for access from multiple users and request for new services for its applications have also increased significantly. In a result there is a rise in packet loss rates and drop in network efficiency. In addition, the inability to support new services has severely hindered the widespread deployment of bandwidth-sensitive applications. So the success of TCP / IP depends on its ability to deliver services in time of extremely high demand and hence various congestion control mechanisms ensure this ability to prosper. The idea behind TCP congestion control is to control network load by having sources adjust their rates according to the level of congestion in the network. This paper focuses on how congestion control and queue management techniques have evolved in due course of time and being modified to minimize the rate of packet loss. Considering RED being most important of all we have optimized the algorithm by challenging the linearity of marking/dropping probability. We conduct a survey by applying different possible functions in contrary to the linear behavior of dropping probability for evaluating their performance in comparison to RED. We implement the optimized RED by simulating a multicast network for TCP flows.

Extended Queue Backward Congestion Control Algorithms for Web Transfers

Rapid development of telecommunication technologies and the ever growing network users demands have made network congestion a prominent problem in today's Internet. Congestion brings significant performance degradation to the network the Quality-of-Service (QoS). The Active Queue Management algorithms (AQM) are used to reduce congestion. In this paper, two AQM algorithms Extended Queue Management Backward Congestion Control Algorithm (EQMBCCA) and QMBCCA, which we have proposed earlier are taken in to consideration. QMBCCA makes use of ISQ (Internet Control Messaging Protocol (ICMP) Source Quench) signaling mechanism and also the CE (Congestion Experienced) bit whenever the average queue size crosses minimum threshold value. Both of them are redundantly used even for mild congestions. EQMBCCA introduces a configurable intermediate threshold value IntThres between the minimum and maximum values and generates ISQ signals only if the congestion crosses this threshold value. Therefore the generation of ISQ messages is significantly reduced here. There is also reliability since CE bits are also set in the packets once congestion occurs. The two algorithms are compared in terms of ISQ traffic, HTTP packet loss, fairness and average web object transfer delay for short lived web transfers. We have found that the performance of EQMBCCA is almost equal to that of QMBCCA and there is a significant reduction in the ISQ traffic in the reverse direction.

Survey on Random Early Detection Mechanism and Its Variants

Active Queue Management (AQM) algorithms are an Implementing schemes, So that packets are transmitted with higher priority than others. Random Early Detection (RED) is the first active queue management algorithm proposed for deployment in TCP/IP networks. RED has some parameter tuning issues that need to be carefully addressed for it to give good performance under different network scenarios. Various algorithms come from RED such as Stabilized RED (SRED), Dynamic RED (DRED), Adaptive RED (ARED) and Flow RED (FRED) these algorithms control congestion by discarding packets with a load dependent probability whenever a queue in the network appear to be congested, this paper will introduced some features about RED and its variants.

Stochastic hyperbolic random early detection controller based on probability density function estimator and particle swarm optimisation algorithm

Most conventional active queue management (AQM) methods are established on the first stochastic moment (mean value) of queue length variation. Yet, these conventional methods have been inefficient so far. Regarding the stochastic attributes of communication networks, AQM methods based on stochastic theory perform better in comparison with conventional AQM methods. A new stochastic AQM advances is advised for the control of queue length probability density function (PDF) for transmission control protocol (TCP). The Parzen window estimate of PDFs exploitation the kernel function is used to constitute the queue PDFs of the TCP/AQM routers. This study proposes a novel stochastic AQM algorithm called hyperbolic random early detection (PDF-HRED) as an appendage of random early detection (RED). The PDF-HRED is based on a stochastic PDF estimation, an HRED controller and particle swarm optimisation (PSO). The numerical simulation results by NS 2 demonstrate that the PDF-HRED controller outperforms the existing AQM schemes with respect to performance characteristics.

CARED: Cautious Adaptive RED gateways for TCP/IP networks

Random Early Detection (RED) is a widely deployed active queue management algorithm that improves the overall performance of the network in terms of throughput and delay. The effectiveness of RED algorithm, however, highly depends on appropriate setting of its parameters. Moreover, the performance of RED is quite sensitive to abrupt changes in the traffic load. In this paper, we propose a Cautious Adaptive Random Early Detection (CARED) algorithm that dynamically varies maximum drop probability based on the level of traffic load to improve the overall performance of the network. Based on extensive simulations conducted using Network Simulator-2 (ns-2), we show that CARED algorithm reduces the packet drop rate and achieves high throughput as compared to RED, Adaptive RED and Refined Adaptive RED. Unlike other RED based algorithms, CARED algorithm does not introduce new parameters to achieve performance gain and hence can be deployed without any additional complexity.