A stochastic approximation approach to active queue management

Bhatnagar, Shalabh; Patel, Sanjeev; Karmeshu,

doi:10.1007/s11235-017-0377-1

Cited by 14 publications

(16 citation statements)

References 45 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Particularly, to reduce packet loss, Effective Random Early Detection (ERED) [6], Adaptive Queue Random Early Detection (AQMRD) [17], Research on Improving Random Early Detection (S-RED) [4], Enhanced RED (EnRED) and Time-window Augmented RED (Windowed-RED) [8]were proposed. Similarly, various methods were proposed to reduce packet droppings, such as Cloudbased Random Early Detection (CRED) [5], Adaptive Threshold RED [11], and Adaptive AQMRD [18]. For a fair resource allocation objective, which implemented by dropping packets based on the shared bandwidth consumed by each flow, Stabilized RED (SRED) [19] and Probability-based RED (P-RED) [20] methods were proposed.…”

Section: Related Workmentioning

confidence: 99%

“…For a fair resource allocation objective, which implemented by dropping packets based on the shared bandwidth consumed by each flow, Stabilized RED (SRED) [19] and Probability-based RED (P-RED) [20] methods were proposed. Various methods were proposed for adaptive parameterestimation, such as: AQMRD [17], CRED [5], Adaptive Threshold RED [11], Adaptive-AQMRD [18], Adaptive Tuning of Drop-Tail (ADT) [21], Adaptive BLUE [22], Self-Tuning Price-based Congestion Control (SPC) [7], New Adaptive RED (NARED) [23] and Queue-Threshold RED (LTRED) [12]. To ease the parameter sensitivity, various AQM methods used fuzzy logic with queue management.…”

Section: Related Workmentioning

confidence: 99%

“…Methods To reduce packet loss ERED [6], AQMRD [17], S-RED[4]EnRED) and Windowed-RED [8]. To reduce packet dropping CRED [5], Adaptive Threshold RED [11], and Adaptive AQMRD [18]. To implement fair resource allocation SRED [19] and P-RED [20].…”

Section: Goalmentioning

confidence: 99%

“…To adaptively estimate the parameter settings AQMRD [17], CRED [5], Adaptive Threshold RED [11], Adaptive-AQMRD [18], Adaptive Tuning of Drop-Tail (ADT) [21], Adaptive BLUE [22], SPC [7], NARED [23] and LTRED [12]. To ease the parameter sensitivity FuzzyRED [13], DeepBlue [24], Fuzzy logic for GRED (GREDFL) [2], and FLRED [3].…”

Section: Goalmentioning

confidence: 99%

See 3 more Smart Citations

Controlling Delay at the Router Buffer using Modified Random Early Detection

Abu-Shareha¹

2019

IJCNC

View full text Add to dashboard Cite

Active Queue Management (AQM) methods are used to manage the buffer of the network routers and avoid the problems caused by network congestion, especially packet loss. Among various AQM methods that have been proposed in the literature, Random Early Detection (RED)method has proved to stabilize the network performance under various traffic loads. However, as the primary concern of RED is to avoid loss when the router buffer overflowed, RED harms the delay at the router and increases the latency. Given that reducing delay is critical to some applications, such as online conferencing and broadcasting, RED needs to be adjusted to ease the delay problem. In this paper, RED is improved by monitoring the delay at the router buffer and implementing packet dropping to handle the issue of network delay and enhance the network performance. Accordingly, the modified method calculates and used a delay parameter with the RED to reduce the delay while maintaining the desirable RED's characteristics. The experimental results showed that the proposed Delay-Controller Random Early Detection (DcRED) improved network performance under various traffic loads. Compared to RED, DcRED results in less delay, while maintaining the loss and dropping rates.

show abstract

Section: Related Workmentioning

confidence: 99%

Section: Related Workmentioning

confidence: 99%

Section: Goalmentioning

confidence: 99%

Section: Goalmentioning

confidence: 99%

See 2 more Smart Citations

Controlling Delay at the Router Buffer using Modified Random Early Detection

Abu-Shareha¹

2019

IJCNC

View full text Add to dashboard Cite

show abstract

“…The models describe that the max p parameter depends on the number of connections, link capacity, RTTs. The upper bound for these two models has been derived in (17) and (18), respectively. Let the mean RTT, number of FTP sources, and link capacity be R, N, and C, respectively:…”

Section: Maximum Probability (Max P )mentioning

confidence: 99%

Sensitivity analysis of queue‐based AQM over network parameters

Patel

2019

IET Networks

Self Cite

View full text Add to dashboard Cite

The comparative analysis of the active queue management (AQM) has been seen in the past work, while the comparative analyses for combined multiple output responses or performance parameters viz., throughput, delay, and loss rate are rarely seen in the literature. The comparative analysis using single variable comprising the output performance parameters has been attempted. The sensitivity of the network parameters over the throughput, delay, and loss rate for different AQM schemes has been measured. An attempt has also been made to stabilise the AQM algorithm that aims to achieve a low-loss rate and a high throughput. The models for the stabilisation of queue size of random early detection (RED) based on the network parameter max p have been studied. The comparative performance analysis of existing stabilised models has also been presented with author's modified RED. The key idea is to modify the existing RED algorithm to achieve stabilisation in queue length at routers with the reduced loss rate as compared to RED.

show abstract

A self-tuning controller for queuing delay regulation in TCP/AQM networks

Kahe

Jahangir

2018

Telecommun Syst

View full text Add to dashboard Cite

AQM router aims primarily to control the network congestion through marking/dropping packets which are used as congestion feedback in traffic sources to balance their flow rate. However, stabilizing queuing delay and maximizing link utilization have been considered as the main control objectives, especially in media dominated networks. Usually, most of the AQM algorithms are designed for a nominal operating point. However, time-varying nature of network parameters frequently violates their robustness bounds. In this paper, a self-tuning compensated PID controller is proposed to address the time-varying nature of network conditions caused by parameter variations and unresponsive connections. The proposed scheme consists of network parameter estimation and a self-tuning AQM. Traffic load, network delay, and bottleneck link capacity are the time-varying network parameters whose variation effects should be compensated by the controller gains adaptation. As the controller gains are simply and directly obtained from the dynamic model, the obtained self-tuning controller can reasonably adapt itself to different operating conditions, while preserving the simplicity of the PI controllers. Packet-level simulations using ns2 show the outperformance of the developed controller for both latency regulation and resource utilization.

show abstract

A stochastic approximation approach to active queue management

Cited by 14 publications

References 45 publications

Controlling Delay at the Router Buffer using Modified Random Early Detection

Controlling Delay at the Router Buffer using Modified Random Early Detection

Sensitivity analysis of queue‐based AQM over network parameters

A self-tuning controller for queuing delay regulation in TCP/AQM networks

Contact Info

Product

Resources

About