A QoS Optimization Technique with Deep Reinforcement Learning in SDN-Based IoT

Moslehi, Mohammadreza; Ebrahimpor-Komleh, Hossein; Goli, Salman; Taji, Reza

doi:10.52547/mjee.15.3.105

Cited by 5 publications

(5 citation statements)

References 14 publications

(24 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In another case of maximizing profit rewards, an investigation of machine learning and deep reinforcement learning algorithms for resource allocation and QoS improvement in SDN-based IoT was conducted. Based on this, a deep reinforcement learning algorithm was proposed to solve the resource allocation of clients/servers in the SDN-based IoT environment, and its performance was verified compared to existing random and round-robin methods [65]. Additionally, a study on intelligent QoS optimization proposed and evaluated an intelligent QoS optimization and network traffic scheduling solution.…”

Section: Evaluation and Methods Analysis Of Reinforcement Learning-ba...mentioning

confidence: 99%

“…Through this, applying reinforcement learning to training networks will likely enable efficient resource management for dynamic (e.g., real-time) changes, making it feasible to maximize tenant benefits and ensure QoS. A study on maximizing cumulative rewards [65] collected information from the SDN controller, and performed network QoS maximization and accumulation of maximum rewards through network state learning and continuous updates. This is an applicable evaluation method for dynamically changing training networks, and will likely make it possible to dynamically manage network traffic and improve efficiency.…”

Section: Evaluation and Methods Analysis Of Reinforcement Learning-ba...mentioning

confidence: 99%

“…These were based on the fact that conventional network slicing does not consider low resource utilization or user QoS due to changes in network demand. Additionally, given the current need for new technology to easily connect heterogeneous devices in heterogeneous network environments due to the exponential growth of communication network devices, considerations for various levels of QoS and policies have been required depending on device type and location in such network and communication environments [65]. Network traffic classification technology through the DRL algorithm has caused issues where the path obtained from single routing metrics like routing hops and link cost may not be the optimal path, and the same link may be selected as the optimal path due to multiple data flows.…”

Section: Reinforcement Learning-based Network Traffic Classification ...mentioning

confidence: 99%

“…There are several considerations for applying such a dynamic resource adjustment algorithm to training networks; it is necessary to account for the impact of factors like initialization, action determination, and learning rate setting on the performance of the dynamic resource adjustment algorithm. A study on QoS and policies [65] proposed solutions for effectively performing traffic classification and QoS management in the network by adjusting rewards and policies through proper agent actions and monitoring QoS parameters. To apply this to training networks, the algorithm, configured in multiple stages based on dynamic management ease, cost efficiency, and adaptability, should enable the agent to take appropriate action, and considerations for monitoring and evaluating QoS parameters are needed.…”

Section: Reinforcement Learning-based Network Traffic Classification ...mentioning

confidence: 99%

See 3 more Smart Citations

An Investigation of Learning Model Technologies for Network Traffic Classification Design in Cyber Security Exercises

Jang,

Kim,

Shin

et al. 2023

IEEE Access

View full text Add to dashboard Cite

With the proliferation of network systems, the boundaries between cyber and physical environments are blurring, leading to an increased risk of sophisticated cyber-attacks equipped with advanced technologies. In particular, as advancements in artificial intelligence through learning models have led to automated attacks and attack scenarios, countries are implementing cyber training and constructing training systems to respond to cyber security threats. This cyber training is based on existing cyber-attacks and conducted in virtual spaces similar to reality, generating network traffic through simulators and focusing on training for attack response and cyber resilience. However, the exponential increase in the number of networkbased devices and the amount of network traffic they generate is leading to a gradual increase in threats to cyber security. In this study, we first investigated the existing port number-based network traffic classification technologies and payload-based network traffic classification technologies to identify their shortcomings in the current network environment. We then categorized existing studies into supervised, unsupervised, and reinforcement learning to analyze the technology of classifying network traffic based on learning models as well as classification methods, procedures, performance standards, evaluation methods, quality of service/quality of experience, etc. Based on the analysis, we presented limitations for application to training networks according to the learning method and suggested recommendations for establishing future research directions. Therefore, refining learning model-based network traffic classification technology will contribute to the construction of automated cyber training grounds such as cyber-attack-defense scenarios, network traffic anomaly detection, and maximizing cumulative rewards.

show abstract

Section: Evaluation and Methods Analysis Of Reinforcement Learning-ba...mentioning

confidence: 99%

Section: Evaluation and Methods Analysis Of Reinforcement Learning-ba...mentioning

confidence: 99%

Section: Reinforcement Learning-based Network Traffic Classification ...mentioning

confidence: 99%

Section: Reinforcement Learning-based Network Traffic Classification ...mentioning

confidence: 99%

See 2 more Smart Citations

An Investigation of Learning Model Technologies for Network Traffic Classification Design in Cyber Security Exercises

Jang,

Kim,

Shin

et al. 2023

IEEE Access

View full text Add to dashboard Cite

show abstract

“…Such a balance aids in evading the pitfalls of local optima and improving convergence speeds. In the research presented by [25], deep reinforcement learning techniques were applied to resource scheduling within the control plane of SDN. The advanced algorithm was developed with the aim of enhancing resource distribution, and it successfully exhibited superior network performance.…”

Section: Related Workmentioning

confidence: 99%

DQQS: Deep Reinforcement Learning-Based Technique for Enhancing Security and Performance in SDN-IoT Environments

Zabeehullah,

Arif,

Ali Khan

et al. 2024

IEEE Access

View full text Add to dashboard Cite

The Internet of Things (IoT) is an emerging technology that allow smart devices to communicate through various heterogeneous channels (wired or wireless). However, for conventional networks, it has become a challenging task to efficiently control and manage the data flows of a huge number of devices. Software-defined networking (SDN) is a new way of thinking about networking. Because it is programmable, flexible, agile, and gives you a big picture of the network, it has tried to solve some IoT problems, like scalability, heterogeneity, and complexity. In large-scale SDN-IoT networks, there is a requirement for routing protocols that are both efficient and secure in order to ensure a superior level of quality of service (QoS) and quality of experience (QoE). To address the above stated challenges, a novel deep reinforcement learning (DRL) known as DQQS model is proposed. The aim is to achieve QoS and QoE while also ensuring the security of the SDN-IoT network. The proposed DQQS model dynamically extracts patterns from the past network history by interacting with the underlying network and generating optimized routing policies. This article employs three network metrics-throughput, latency, and the probability of avoiding malicious nodes-to measure the performance of DQQS. Simulations reveal that the proposed framework outperforms four state-of-the-art routing algorithms: OSPF, L-L Routing, Sailfish Routing, and RL-Routing in terms of both throughput and latency. For instance, in an attacked environment, the proposed DQQS model achieved the highest throughput value of 14.5 Mbps, surpassing OSPF at 8 Mbps, L-L at 8.2 Mbps, Sailfish at 9 Mbps, and RL at 9.5 Mbps. Similarly, this model exhibited superior performance in latency, recording the lowest latency value of 52 ms, compared to OSPF 88 ms, L-L 85 ms, Sailfish 72 ms, and RL 75 ms routing algorithms. The experimental results demonstrate that this new DQQS model is a pioneering deep reinforcement learning-based technique that optimally addresses secure routing in the SDN-IoT environment, ensuring enhanced quality of service and experience, and outperforming state-of-the art DL methodologies in both security and network performance metrics.

show abstract

Unlocking QoS Potential: Integrating IoT services and Monte Carlo Control for heterogeneous IoT device management in gateways

Chakour,

Mhammedi,

Daoui

et al. 2024

Computer Networks

View full text Add to dashboard Cite

A QoS Optimization Technique with Deep Reinforcement Learning in SDN-Based IoT

Cited by 5 publications

References 14 publications

An Investigation of Learning Model Technologies for Network Traffic Classification Design in Cyber Security Exercises

An Investigation of Learning Model Technologies for Network Traffic Classification Design in Cyber Security Exercises

DQQS: Deep Reinforcement Learning-Based Technique for Enhancing Security and Performance in SDN-IoT Environments

Unlocking QoS Potential: Integrating IoT services and Monte Carlo Control for heterogeneous IoT device management in gateways

Contact Info

Product

Resources

About