An adaptive backoff selection scheme based on Q-learning for CSMA/CA

Zheng, Zhichao; Jiang, Shengming; Feng, Ruoyu; Ge, Lige; Gu, Chongchong

doi:10.1007/s11276-023-03257-0

Cited by 7 publications

(4 citation statements)

References 27 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Their approach integrates Q-learning to effectively identify the optimal backoff values, thereby substantially reducing the frequency of collisions. In addition, Zheng et al [17] suggested a Qlearning scheme modification to improve the performance of the MAC protocol by adjusting the learning rate and exploring different 𝜀-greedy exploration rates.…”

Section: Related Workmentioning

confidence: 99%

“…In the field of machine learning (ML), reinforcement learning (RL) [11] is particularly influential for tasks like choosing optimal CW values. Key researchers in this field, i.e., Kim & Hwang [12], Zerguine et al [13], Kwon et al [14], Pan et al [15], Lee et al [16], and Zheng et al [17] have leveraged the Q-learning algorithm, as a core of the fundamental RL mechanism, to determine the appropriate CW value according to the consecutive successes or collisions transmitted packets. Nevertheless, Q-learning has the drawback which expensive for the agent, whereas in the earlier stages of the learning phase, each pair of states and actions must be exhaustively explored to converge towards the optimal policy.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Deep reinforcement learning-based contention window optimization for IEEE 802.11 networks

Tu,

Ma,

2024

Preprint

View full text Add to dashboard Cite

This study focuses on optimizing the contention window (CW) in IEEE 802.11 networks using deep reinforcement learning (DRL) to enhance the effectiveness of the contention mechanism. Recent research has employed a deep Q-learning network (DQN) as one type of DRL for CW size selection tasks to maximize network throughput. However, a notable limitation of DQN is the substantial overestimation error, which means the predicted reward value significantly deviates from the actual value. To address this issue, our study introduces the smart exponential-threshold-linear with double deep Q-learning network (SETL-DDQN) in a wireless networks scenario, with the aim to mitigate the overestimation error via the CW threshold size optimization with the help of a DDQN-based approach during the learning phase. We experimented with our proposed SETL-DDQN in both static and dynamic scenarios and conducted an analysis to solve the overestimation problem, then enhance the long-term simulation stability. Our experimental results demonstrate that SETL-DDQN achieves more efficient packet transmissions than related existing mechanisms.

show abstract

Section: Related Workmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Deep reinforcement learning-based contention window optimization for IEEE 802.11 networks

Tu,

Ma,

2024

Preprint

View full text Add to dashboard Cite

show abstract

“…Zheng, Z. et al [29] When data are passed through an intermediary, there is always the risk that it could be intercepted or that a third party could gain access to it. When devices communicate directly with each other, that risk is eliminated.…”

Section: Alani T O Et Al [19]mentioning

confidence: 99%

“…This can be especially helpful when dealing with large amounts of data, or when time is of the essence. Zheng, Z. et al [29] expressed that an advantage of device-to-device communication is that it can be more secure than other methods. When data are passed through an intermediary, there is always the risk that it could be intercepted or that a third party could gain access to it.…”

Section: Introductionmentioning

confidence: 99%

Smart Load-Based Resource Optimization Model to Enhance the Performance of Device-to-Device Communication in 5G-WPAN

Logeshwaran¹,

Kiruthiga²,

Kannadasan

et al. 2023

Electronics

View full text Add to dashboard Cite

In wireless personal area networks (WPANs), devices can communicate with each other without relying on a central router or access point. They can improve performance and efficiency by allowing devices to share resources directly; however, managing resource allocation and optimizing communication between devices can be challenging. This paper proposes an intelligent load-based resource optimization model to enhance the performance of device-to-device communication in 5G-WPAN. Intelligent load-based resource optimization in device-to-device communication is a strategy used to maximize the efficiency and effectiveness of resource usage in device-to-device (D2D) communications. This optimization strategy is based on optimizing the network’s resource load by managing resource utilization and ensuring that the network is not overloaded. It is achieved by monitoring the current load on the network and then adjusting the usage of resources, such as bandwidth and power, to optimize the overall performance. This type of optimization is essential in D2D communication since it can help reduce costs and improve the system’s performance. The proposed model has achieved 86.00% network efficiency, 93.74% throughput, 91.94% reduced latency, and 92.85% scalability.

show abstract

Best-Effort Networking Systems

Jiang

2024

Lecture Notes in Networks and Systems

View full text Add to dashboard Cite

An adaptive backoff selection scheme based on Q-learning for CSMA/CA

Cited by 7 publications

References 27 publications

Deep reinforcement learning-based contention window optimization for IEEE 802.11 networks

Deep reinforcement learning-based contention window optimization for IEEE 802.11 networks

Smart Load-Based Resource Optimization Model to Enhance the Performance of Device-to-Device Communication in 5G-WPAN

Best-Effort Networking Systems

Contact Info

Product

Resources

About