Active Queue Management Methods Research Articles

A four-state Markov model for modelling bursty traffic and benchmarking of random early detection

Active Queue Management (AQM) techniques are crucial for managing packet transmission efficiently, maintaining network performance, and preventing congestion in routers. However, achieving these objectives demands precise traffic modeling and simulations in extreme and unstable conditions. The internet traffic has distinct characteristics, such as aggregation, burstiness, and correlation. This paper presents an innovative approach for modeling internet traffic, addressing the limitations of conventional modeling and conventional AQM methods' development, which are primarily designed to stabilize the network traffic. The proposed model leverages the power of multiple Markov Modulated Bernoulli Processes (MMBPs) to tackle the challenges of traffic modeling and AQM development. Multiple states with varying probabilities are used to model packet arrivals, thus capturing the burstiness inherent in internet traffic. Yet, the overall probability is maintained identical, irrespective of the number of states (one, two, or four), by solving linear equations with multiple variables. Random Early Detection (RED) was used as a case study method with different packet arrival probabilities based on MMBPs with one, two, and four states. The results showed that the proposed model influences the outcomes of AQM methods. Furthermore, it was found that RED might not effectively address network burstiness due to its relatively slow reaction time. As a result, it can be concluded that RED performs optimally only with a single-state model.

Fuzzy Comprehensive Random Early Detection of Router Congestion

The queue length and the load rate should be monitored to overcome the problem of router congestion due to the increase in network utilization and achieve a high-speed transmission. Previous active queue management methods manage the queued packets in the router buffer to maintain high network performance. However, these methods depend on monitoring indicators that do not cover all the congestion signs, leading to packet loss and delay. Accordingly, all the congestion signs should be wrapped into these indicators and managed by an algorithm that randomly drops packets to avoid global synchronization, loss, and delay. In this paper, a fuzzy comprehensive random early detection (FCRED) is proposed to deal with the gap in network monitoring and congestion control at the router buffer. FCRED is built by using three indicators, which monitor the router's arrival, departure, and queue length. Accordingly, a fuzzy inference process is developed to manage these indicators and calculate the dropping probability (Dp). Simulation results show that FCRED improves loss and packet dropping under various network statuses compared with RED, BLUE, and ERED. In terms of loss, FCRED achieves zero loss at high congested status. For dropping, FCRED achieves an optimal rate of 0.47 with an arrival rate of 0.95. For the throughput and delay, FCRED achieves the best results. Accordingly, the proposed FCRED method achieves zero loss and reduces packet dropping from 0.28 to 0.21, a 25% reduction compared with the best performance of these methods. Compared with recent fuzzy-based methods, the proposed FCRED achieves comparable results and outperforms them by dropping more packets to avoid loss, which in such case is necessary dropping.

Linear Random Early Detection for Congestion Control at the Router Buffer

Active Queue Management (AQM) methods control the router's buffer to maintain high network performance and control congestion at the router buffer. Random Early Detection (RED) method is the most well-known and the most utilized AQM. RED suffers from a high dropping rate, which motivates the later AQM methods to use more complex processes, which reach the limits of using fuzzy systems as a processing technique. Yet, high computational cost affects the performance at the router specifically and the network as a whole, with so-called processing delay. In this paper, a linear version of RED (LRED) is presented to reduce the computational cost of the original RED and maintain the network performance in terms of throughput, delay, dropping, and loss. LRED is built based on two distinctive features, simplifying the congestion indicator calculation and reducing the operations in calculating the dropping probability. The experimental results showed that the proposed method reduces the delay and the processing time while maintaining the throughput and loss of the RED method.

Evaluation of Active Queue Management Methods based on Integrated AHP, TOPSIS and Fuzzy TOPSIS

The Evaluation and selection of Active Queue Management (AQM) methods are a complicated and challenging task. Improper AQM method can cause up-mark and network malfunctioning. To achieve satisfactory performance, various evaluation criteria need to be considered. In order to find the limitations of how the criteria are determined, there is a need to find out how their procedures change according to the evaluation and benchmarking process of AQM. This article focuses on the evaluation and benchmarking of Active Queue Management methods using Multi-Criteria Decision-Making (MCDM) techniques. MCDM uses different techniques to figure out the best alternative from multi-criteria and multi alternative conditions. Analytic Hierarchy Process (AHP) is a multi-criteria decision-making technique that is used to assign weights to criteria. Technique for Order of Preference by Similarity to Ideal Solution (TOPSIS) is used for ranking and selection of different alternatives by using distance measures. Whereas, FUZZY TOPSIS is used for criteria weightage and alternatives rating. MCDM calculation is performed on AQM methods and it is tried to discuss all these techniques. MCDM uses performance, overhead, and configuration criteria to evaluate and select the best AQM method that helps to determine solutions for future directions. The results show that Random Early Detection (RED) method got higher ranking score records as compared to other AQM methods.

Fuzzy-Logic Based Active Queue Management Using Performance Metrics Mapping into Multi-Congestion Indicators

Abstract The congestion problem at the router buffer leads to serious consequences on network performance. Active Queue Management (AQM) has been developed to react to any possible congestion at the router buffer at an early stage. The limitation of the existing fuzzy-based AQM is the utilization of indicators that do not address all the performance criteria and quality of services required. In this paper, a new method for active queue management is proposed based on using the fuzzy logic and multiple performance indicators that are extracted from the network performance metrics. These indicators are queue length, delta queue and expected loss. The simulation of the proposed method show that in high traffic load, the proposed method preserves packet loss, drop packet only when it is necessary and produce a satisfactory delay that outperformed the state-of-the-art AQM methods.

Benchmarking of AQM methods of network congestion control based on extension of interval type-2 trapezoidal fuzzy decision by opinion score method

This study presents a benchmarking and evaluation approach for active queue management (AQM) network congestion control methods, which are considered as a problem of multi-criteria decision-making (MCDM). In recent years, the development of MCDM methods has been studied from various perspectives. The latest one called fuzzy decision by opinion score method (FDOSM) has proved its efficiency in solving the concerns faced by other methods. However, the approach of FDOSM and its extension is based on fuzzy type-1, which suffers from issues, especially minimising the effect of data uncertainties. Therefore, this study extended FDOSM into a fuzzy type-2 environment that utilises interval type-2 trapezoidal (IT2T) membership, and then discusses the effectiveness of such membership on AQM method benchmarking. The methodology of this study involves two consecutive phases. The first phase is the construction of a decision matrix utilised in AQM method benchmarking based on a list of AQM methods and multiple evaluation criteria. The second phase is regarding the new method (IT2T-FDOSM), which illustrated two main stages, namely, data transformation unit and data processing. The findings of this study are the following: (1) Individual results of benchmarking which used six decision-makers are almost similar, with the AQM fuzzy GRED method ranked as the best. (2) The group benchmarking results show that a relatively similar order and fuzzy GRED method is the best as well. (3) IT2T-FDOSM can deal with the uncertainty problem properly. (4) The results show significant differences amongst the groups’ scores, which indicate the validity of the IT2T-FDOSM results.

Multidimensional Benchmarking Framework for AQMs of Network Congestion Control Based on AHP and Group-TOPSIS

This paper aims to propose a grouping framework for benchmarking the active queue management (AQM) methods of network congestion control based on multicriteria decision-making (MCDM) techniques to assist developers of AQM methods in selecting the best AQM method. Given the current rapid development of the AQM techniques, determining which of these algorithms is better than the other is difficult because each algorithm performs better in a specific metric(s). Current benchmarking studies benchmark the AQM methods from a single incomplete prospective. In each proposed AQM method, the benchmarking was achieved with reference to some evaluation measures that are relatively close to the desired goal being followed during the development of the AQM methods. Furthermore, the benchmarking frameworks of AQM methods are complicated and challenging because of the following reasons: (1) the technical details of the AQM methods are adapted and the input parameters are selected according to the sensitivity of the AQM methods; and (2) a framework is developed and designed for simulating AQM methods, the simulated network and the collected results. For this purpose, a set of criteria for AQM comparison are determined. These criteria are performance, processing overhead and configuration. The benchmarking framework is developed based on the crossover of three groups of multi-evaluation criteria and several AQM methods as a proof of concept. The AQM families that are implemented and utilized in experiments to generate the data that are used as a proof of concept of our proposed framework are the parameter-based (pars) and fuzzy-based AQM methods. Accordingly, constructing the decision matrix (DM) that will be used to generate the final results is necessary. Subsequently, the underlying AQM methods are benchmarked and ranked using MCDM techniques, namely, integrated analytical hierarchy process (AHP) and technique for order of preference by similarity to ideal solution (TOPSIS). The validation was performed objectively. The [Formula: see text] deviation was computed to ensure that the AQM methods ranking undergo systematic ranking. Results illustrate that (1) the integration of AHP and TOPSIS solves the AQM method benchmarking problems; (2) results of the individual TOPSIS context clearly show variances among the ranking results of the six experts; (3) the ranks of the AQM methods obtained from internal and external TOPSIS group decision-making are nearly similar, with random early detection method being ranked as the best one; and (4) in the objective validation, significant differences were found between the groups’ scores, thereby indicating that the ranking results of internal and external TOPSIS group decision-making were valid.

Multi-indicator Active Queue Management Method

A considerable number of applications are running over IP networks. This increased the contention on the network resource, which ultimately results in congestion. Active queue management (AQM) aims to reduce the serious consequences of network congestion in the router buffer and its negative effects on network performance. AQM methods implement different techniques in accordance with congestion indicators, such as queue length and average queue length. The performance of the network is evaluated using delay, loss, and throughput. The gap between congestion indicators and network performance measurements leads to the decline in network performance. In this study, delay and loss predictions are used as congestion indicators in a novel stochastic approach for AQM. The proposed method estimates the congestion in the router buffer and then uses the indicators to calculate the dropping probability, which is responsible for managing the router buffer. The experimental results, based on two sets of experiments, have shown that the proposed method outperformed the existing benchmark algorithms including RED, ERED and BLUE algorithms. For instance, in the first experiment, the proposed method resides in the third-place in terms of delay when compared to the benchmark algorithms. In addition, the proposed method outperformed the benchmark algorithms in terms of packet loss, packet dropping, and packet retransmission. Overall, the proposed method outperformed the benchmark algorithms because it preserves packet loss while maintaining reasonable queuing delay.

Multidimensional benchmarking of the active queue management methods of network congestion control based on extension of fuzzy decision by opinion score method

Multidimensional benchmarking of the active queue management methods of network congestion control based on extension of fuzzy decision by opinion score method

An Exponential Active Queue Management Method Based on Random Early Detection

Congestion is a key topic in computer networks that has been studied extensively by scholars due to its direct impact on a network’s performance. One of the extensively investigated congestion control techniques is random early detection (RED). To sustain RED’s performance to obtain the desired results, scholars usually tune the input parameters, especially the maximum packet dropping probability, into specific value(s). Unfortunately, setting up this parameter into these values leads to good, yet biased, performance results. In this paper, the RED-Exponential Technique (RED_E) is proposed to deal with this issue by dropping arriving packets in an exponential manner without utilizing the maximum packet dropping probability. Simulation tests aiming to contrast E_RED with other Active Queue Management (AQM) methods were conducted using different evaluation performance metrics including mean queue length (mql), throughput (T), average queuing delay (D), overflow packet loss probability (PL), and packet dropping probability (DP). The reported results showed that E_RED offered a marginally higher satisfactory performance with reference to mql and D than that found in common AQM methods in cases of heavy congestion. Moreover, RED_E compares well with the considered AQM methods with reference to the above evaluation performance measures using minimum threshold position (min threshold) at a router buffer.

Multi-Criteria Evaluation and Benchmarking for Active Queue Management Methods: Open Issues, Challenges and Recommended Pathway Solutions

The evaluation and benchmarking processes of active queue management (AQM) methods are complicated and challenging. Several evaluation criteria/metrics must be considered before an AQM method can yield satisfactory performance using specific metric(s). Further investigations are required to highlight the limitations of how criteria/metrics are determined and how their procedures accord with the evaluation and benchmarking processes of AQM. In this paper, we presented comprehensive insights into the multi-criteria evaluation and benchmarking of AQM methods based on two critical directions. First, current AQM evaluation criteria are collected, analyzed and categorized. Second, these AQM evaluation criteria highlight conflicting issues and benchmarking techniques to identify weak points, and possible solutions are discussed. The findings of this study are as follows: (1) The limitations and problems of existing AQM evaluation and benchmarking methods, such as multi-evaluation criteria, criteria trade-off, benchmarking and criteria significance, are presented and emphasized. (2) Multi-criteria decision-making using multiple criteria, such as performance, processing overhead and configuration, can be used to benchmark numerous AQM methods to determine solutions for future directions.

IGRED: An Improved Gentle Random Early Detection Method for Management of Congested Networks

Controlling congested router buffers of a network has a crucial role in improving network’s performance. This paper proposes a novel Active Queue Management (AQM) method named Improved Gentle Random Early Detection (IGRED) that based on GRED algorithm, which counted as one of the popular AQM methods. The proposed is mainly developed to overcome the problems faced with classic GRED. The initial packet-dropping probability depends on several parameters such as the average queue length, maximum value of packet dropping probability, minimum and maximum thresholds, etc. IGRED reduces its reliance on the GRED’s parameters through shrinking these parameters. The results shows, when congestion is taken place, the proposed IGRED provides more satisfactory performance with reference to mean queue length, average queuing delay, and overflow packet loss probability.

Enhanced Random Early Detection using Responsive Congestion Indicators

Random Early Detection (RED) is an Active Queue Management (AQM) method proposed in the early 1990s to reduce the effects of network congestion on the router buffer. Although various AQM methods have extended RED to enhance network performance, RED is still the most commonly utilized method; this is because RED provides stable performance under various network statuses. Indeed, RED maintains a manageable buffer queue length and avoids congestion resulting from an increase in traffic load; this is accomplished using an indicator that reflects the status of the buffer and a stochastic technique for packet dropping. Although RED predicts congestion, reduces packet loss and avoids unnecessary packet dropping, it reacts slowly to an increase in buffer queue length, making it inadequate to detect and react to sudden heavy congestion. Due to the aforementioned limitation, RED is found to be significantly influenced by the way in which the congestion indicator is calculated and used. In this paper, RED is modified to enhance its performance with various network statuses. RED technique is modified to overcome several disadvantages in the original method and enhance network performance. The results indicate that the proposed Enhanced Random Early Detection (EnRED) and Time-window Augmented RED (Windowed-RED) methods—compared to the original RED, ERED and BLUE methods—enhances network performance in terms of loss, dropping and packet delay.

Stabilizing Average Queue Length in Active Queue Management Method

This paper proposes the Stabilized (DGRED) method for congestion detection at the router buffer. This method aims to stabilize the average queue length between allocated minthre_shold and doublemaxthre_shold positions to increase the network performance. The SDGRED method is simulated and compared with Gentle Random Early Detection (GRED) and Dynamic GRED active queue management methods. This comparison is built on different important measures, such as dropping probability, throughput, average delay, packet loss, and mean queue length for packets. The evaluation aims to identify which method presents better simulation performance measurement results when non-congestion or congestion situations occur at the router buffers in congestion control. The results show that at high packet arrival probability, the proposed algorithm helps provide lesser queue length values, delayed time, and packet loss compared with current methods. Furthermore, SDGRED generates adequate throughput at high packet arrival probability.

English

English

Performance study of Active Queue Management methods: Adaptive GRED, REDD, and GRED-Linear analytical model

Congestion control is one of the hot research topics that helps maintain the performance of computer networks. This paper compares three Active Queue Management (AQM) methods, namely, Adaptive Gentle Random Early Detection (Adaptive GRED), Random Early Dynamic Detection (REDD), and GRED Linear analytical model with respect to different performance measures. Adaptive GRED and REDD are implemented based on simulation, whereas GRED Linear is implemented as a discrete-time analytical model. Several performance measures are used to evaluate the effectiveness of the compared methods mainly mean queue length, throughput, average queueing delay, overflow packet loss probability, and packet dropping probability. The ultimate aim is to identify the method that offers the highest satisfactory performance in non-congestion or congestion scenarios. The first comparison results that are based on different packet arrival probability values show that GRED Linear provides better mean queue length; average queueing delay and packet overflow probability than Adaptive GRED and REDD methods in the presence of congestion. Further and using the same evaluation measures, Adaptive GRED offers a more satisfactory performance than REDD when heavy congestion is present. When the finite capacity of queue values varies the GRED Linear model provides the highest satisfactory performance with reference to mean queue length and average queueing delay and all the compared methods provide similar throughput performance. However, when the finite capacity value is large, the compared methods have similar results in regard to probabilities of both packet overflowing and packet dropping.

Improving network congestion: A RED-based FuzzyPID approach

This research proposes a new active queue management method (or FPID) to improve network congestion, which integrates the random early detection (RED) method and fuzzy proportional integral derivative (FuzzyPID) approach. Fuzzy theory applied together with PID helps control the buffer queue within the target. This article provides the literature and theoretical background of network congestion problems, active queue management methods, automatic control systems, and fuzzy controllers. Following the discussion of the proposed FPID method and the experimental design, a simulation test is conducted and analyzed in comparison to the existing popular active queue management methods. The results prove the superiority of the proposed method.

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.

MNEWQUE: A New Approach to TCP/AQM with ECN

In order to improve the performance of congested routers, an Active Queue Management (AQM) is proposed. AQM can potentially reduce packet loss rate in the Internet. This is used to control congestion at the router, where packets are dropped before queue become full. A new framework of AQM, namely Modified NEWQUE (MNEWQUE) active queue management algorithm supporting explicit congestion notification (ECN), is developed by changing constant factor K in NEWQUE AQM. The objective of the new algorithm is to improve performance of congested routers by keeping low queuing delay, packet drop rate low, link utilization high, and link utilization stable. The MNEWQUE AQM is implemented with help of ns2 simulator. The simulation shows that the proposed design outperforms other AQM methods in terms of queuing delay, packet loss, and link utilization.

A Fuzzy Logic Based Network Congestion Control Using Active Queue Management Techniques

Conventional IP routers are passive devices that accept packets and perform the routing function on any input. Usually the tail-drop (TD) strategy is used where the input which exceeds the buffer capacity are simply dropped. In active queue management (AQM) methods routers manage their buffers by dropping packets selectively. We study one of the AQM methods called as random exponential marking (REM). We propose an intelligent approach to AQM based on fuzzy logic controller (FLC) to drop packets dynamically, keep the buffer size around desired level and also prevent buffer overflow. Our proposed approach is based on REM algorithm, which drops the packets by drop probability function. In our proposal we replace the drop probability function by a FLC to drop the packets, stabilize the buffer around the desired size and reduce delay. Simulation results show a better regulation of the buffer. Keywords: Random exponential marking; Active queue management; Fuzzy logic controller; Pro-active queue management. © 2010 JSR Publications. ISSN: 2070-0237 (Print); 2070-0245 (Online). All rights reserved. DOI: 10.3329/jsr.v2i2.2786 J. Sci. Res. 2 (2), 273-284 (2010)