Performance Evaluation of Optimized IEEE 802.11ax in IIoT Environments: Throughput, Latency, and Packet Loss Improvements

Abstract

This paper presents a comprehensive performance evaluation of optimized IEEE 802.11ax networks in Industrial Internet of Things (IIoT) environments focusing on three critical performance metrics: throughput, latency, and packet loss. Using a Deep Reinforcement Learning (DRL)-based optimization strategy, the researchers configured MAC layer parameters to meet the demands of heterogeneous industrial traffic demands. Justification for the use of DRL is provided by its ability to handle complex, stochastic environments effectively.  MATLAB simulations modelled and analyzed network performance under harsh IIoT environments, characterized by electromagnetic interference (EMI), fluctuating traffic loads, high-density device deployments, and significant physical obstructions such as metal structures. The results showed that the optimized network significantly improved system throughput, average transmission delay, and packet retransmission rate. The peak throughput increased from 1420 Mbps to 2150 Mbps, and the highest packet loss ratio decreased from 32.5% to 23%. Latency also saw notable improvement, with the number of nodes experiencing latency greater than 0.1 seconds decreased from 5 to 1. These findings demonstrated that the proposed optimization strategy for IEEE 802.11ax systems can significantly enhance performance in IIoT environments, making them more reliable and efficient for industrial applications.