Frame Size Optimization Using a Machine Learning Approach in WLAN Downlink MU-MIMO Channel

TL;DR

This paper introduces a machine learning method to optimize frame size in WLAN downlink MU-MIMO channels, significantly improving throughput by adapting to traffic and channel conditions.

Contribution

It presents a novel neural network-based approach for frame size optimization that considers traffic heterogeneity and channel variability in WLAN systems.

Findings

Neural network accurately models throughput as a function of frame size.

Optimized frame size improves system throughput under diverse traffic patterns.

Method outperforms traditional FIFO aggregation in various WLAN scenarios.

Abstract

The IEEE 802.11ac/n introduced frame aggregation technology to accommodate the growing traffic demand and increase the performance of transmission efficiency and channel utilization. This is achieved by allowing many packets to be aggregated per transmission which realized a significant enhancement in the throughput performance of WLAN. However, it is difficult to efficiently utilize the benefits of frame aggregation in the downlink MU-MIMO channels as stations have heterogeneous transmission demands and data transmission rates. As a result of this, wasted space channel time will occur which degrades transmission efficiency. In addressing these challenges, the existing studies have proposed different approaches. However, most of these approaches did not consider a machine-Learning based optimization solution. The main contribution of this paper is to propose a machine-learning-based frame size optimization solution to maximize the system throughput of WLAN in the downlink MU-MIMO channel. In this approach, the Access Point (AP) performs the maximum system throughput measurement and collected frame size-system throughput patterns which contain knowledge about the effects of traffic patterns, channel conditions, and number of stations(STAs). Based on these patterns,our approach uses a neural network to correctly model the system throughput as a function of the system frame size. After training the neural network, we obtain the gradient information to adjust the frame size. the performance of the proposed Machine learning(ML) approach is evaluated over the FIFO aggregation algorithm under the effects of heterogenous traffic patterns for VoIP and video applications, channel conditions, and number of stations.