Supervised Learning of Neural Networks for Active Queue Management in the Internet

Szyguła, Jakub; Domański, Adam; Domańska, Joanna; Marek, Dariusz; Filus, Katarzyna; Mendla, Szymon

doi:10.3390/s21154979

Cited by 13 publications

(3 citation statements)

References 50 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The model uses the Gaussian method. The study demonstrates the effectiveness of the active queue management mechanism based on neural networks (Szyguła et al, 2021). In another study, the authors proposed the Federated Intelligence (FIAQM) architecture for AQM, using the Federated Learning approach.…”

Section: Related Workmentioning

confidence: 77%

The Impact of Denial-of-Service Attacks and Queue Management Algorithms on Cellular Networks

ÇAKMAK

2023

jista

View full text Add to dashboard Cite

In today's digital landscape, Distributed Denial of Service (DDoS) attacks stand out as a formidable threat to organisations all over the world. As known technology gradually advances and the proliferation of mobile devices, cellular network operators face pressure to fortify their infrastructure against these risks. DDoS incursions into Cellular Long-Term Evolution (LTE) networks can wreak havoc, elevate packet loss, and suboptimal network performance. Managing the surges in traffic that afflict LTE networks is of paramount importance. Queue management algorithms emerge as a viable solution to wrest control over congestion at the Radio Link Control (RLC) layer within LTE networks. These algorithms work proactively, anticipating, and mitigating congestion by curtailing data transfer rates and fortifying defences against potential DDoS onslaughts. In the paper, we delve into a range of queue management methods Drop-Tail, Random Early Detection (RED), Controlled Delay (CoDel), Proportional Integral Controller Enhanced (PIE), and Packet Limited First In, First Out queue (pFIFO). Our rigorous evaluation of these queue management algorithms hinges on a multifaceted assessment that encompasses vital performance parameters. We gauge the LTE network's resilience against DDoS incursions, measuring performance based on end-to-end delay, throughput, packet delivery rate (PDF), and fairness index values. The crucible for this evaluation is none other than the NS3 simulator, a trusted platform for testing and analysis. The outcomes of our simulations provide illuminating insights. CoDel, RED, PIE, pFIFO, and Drop-Tail algorithms emerge as top performers in succession. These findings underscore the critical role of advanced queue management algorithms in fortifying LTE networks against DDoS attacks, offering robust defences and resilient network performance.

show abstract

Section: Related Workmentioning

confidence: 77%

The Impact of Denial-of-Service Attacks and Queue Management Algorithms on Cellular Networks

ÇAKMAK

2023

jista

View full text Add to dashboard Cite

show abstract

“…COBALT(the CoDel and BLUE Alternate algorithm) [10] is currently used in Linux4. 19. When there is no response flow, COBALT significantly reduces the queue waiting time.…”

Section: Related Workmentioning

confidence: 99%

“…By dealing with the problems of delayed INT information and overreaction to INT information during congestion, HPCC can simultaneously achieve the three properties of ultralow latency, high bandwidth, and stability. Szyguła et al [19] used machine learning to create a model that replicates the behavior of the AQM mechanism, proposed a neural network-based AQM mechanism, and demonstrated its effectiveness. Refs.…”

Section: Related Workmentioning

confidence: 99%

A Sketch-Based Fine-Grained Proportional Integral Queue Management Method

Zhu,

Sun,

Jiang

et al. 2023

Axioms

View full text Add to dashboard Cite

The phenomenon “bufferbloat” occurs when the buffers of the network intermediary nodes fill up, causing long queuing delays. This has a significant negative impact on the quality of service of network applications, particularly those that are sensitive to time delay. Many active queue management (AQM) algorithms have been proposed to overcome this problem. Those AQMs attempt to maintain minimal queuing delays and good throughput by purposefully dropping packets at network intermediary nodes. However, the existing AQM algorithms mostly drop packets randomly based on a certain metric such as queue length or queuing delay, which fails to achieve fine-grained differentiation of data streams. In this paper, we propose a fine-grained sketch-based proportional integral queue management algorithm S-PIE, which uses an additional measurement structure Sketch for packet frequency share judgment based on the existing PIE algorithm for the fine-grained differentiation between data streams and adjust the drop policy for a differentiated packet drop. Experimental results on the NS-3 simulation platform show that the S-PIE algorithm achieves lower average queue length and RTT and higher fairness than PIE, RED, and CoDel algorithms while maintaining a similar throughput performance, maintaining network availability and stability, and improving network quality of service.

show abstract

TCP Congestion Management Using Deep Reinforcement Trained Agent for RED

Ali,

Öztürk

2024

Concurrency and Computation

View full text Add to dashboard Cite

Increasing data transmission volumes are causing more frequent and more severe network congestion. In order to handle spikes in network traffic, a substantially bigger buffer has been included into the system. Bufferbloat, which happens when a bigger buffer is implemented, exacerbates network congestion. Using the transfer control protocol (TCP) congestion management strategy with active queue management (AQM) can fix this issue. As congestion increases, it becomes increasingly difficult to forecast and fine‐tune dynamic AQM/TCP systems in order to achieve acceptable performance. To shed new light on the AQM system, we plan to use deep reinforcement learning (DRL) techniques. It is possible that AQM can learn about the appropriate drop policy the same way people do when using a model‐free technique like DRL‐AQM. After training in a simple network scenario, DRL‐AQM is able to recognize complex patterns in the data traffic model and apply them to improve performance in a wide variety of scenarios. Offline training precedes deployment in our approach. In many cases, the model does not require any further parameter tweaks after training. Even in the most complicated networks, AQM algorithms have proven to be effective, regardless of the network's complexity. Minimizing buffer capacity use is an important goal of DRL‐AQM. It automatically and continually adjusts to changes in network connectivity.

show abstract

Supervised Learning of Neural Networks for Active Queue Management in the Internet

Cited by 13 publications

References 50 publications

The Impact of Denial-of-Service Attacks and Queue Management Algorithms on Cellular Networks

The Impact of Denial-of-Service Attacks and Queue Management Algorithms on Cellular Networks

A Sketch-Based Fine-Grained Proportional Integral Queue Management Method

TCP Congestion Management Using Deep Reinforcement Trained Agent for RED

Contact Info

Product

Resources

About