Predicted Contention Resolution in 802.11ax Based on Reinforcement Learning

Main Article Content

Mohammed AZZA , Amina TAOULI

Abstract

in this paper, we exploits the spatial reuse SR mechanism of IEEE 802.11ax by integrating the color scheme to predict, adapts the OBSS/PD threshold, and adjusts the transmission power level. With the rapid development of modern technology, networks of communication have turned into a complex and colossal system. The new network technology, such as WIFI 6, is committed to providing more great performance and improvements where the density and complexity of the network will more increase and new requirements and problems will appear. In this solution, we propose a prediction reinforcement-learning model in network traffic based on the dynamic exponential smoothing model and optimises the smoothing coefficient of the model through the hyperbolic cosine.  Moreover, our solutions based on the network traffic prediction improve the many parameters such as throughput and de delay. The simulation results show that the advantage of prediction model with optimised hyperbolic cosine has a high accuracy and stability in dense scenario

Downloads

Download data is not yet available.

Metrics

Metrics Loading ...

Article Details

How to Cite
Mohammed AZZA , Amina TAOULI. (2023). Predicted Contention Resolution in 802.11ax Based on Reinforcement Learning. Turkish Journal of Computer and Mathematics Education (TURCOMAT), 14(1), 207–221. https://doi.org/10.17762/turcomat.v14i1.13459
Section
Research Articles