Accurate and Efficient Prediction of Wi-Fi Link Quality Based on Machine Learning

2025-09-23

AI Summary

To address the challenge of highly volatile and difficult-to-predict Wi-Fi link quality in industrial environments, this paper proposes a lightweight, channel-agnostic machine learning prediction model. The model adopts a linear combination architecture based on exponential moving averages, achieving both low computational complexity and high prediction accuracy—making it suitable for resource-constrained embedded hardware. A key innovation is the channel-agnostic training paradigm, enabling cross-device and cross-band generalization while significantly reducing manufacturers’ modeling overhead. Evaluated on a real-world industrial Wi-Fi testbed, the proposed model achieves an average normalized mean squared error (MSE) below 0.08 across diverse operational conditions, outperforming conventional approaches. Moreover, its channel-agnostic variant attains performance comparable to channel-specific models while offering superior deployment flexibility. These advances collectively enhance the reliability and scalability of industrial Wi-Fi systems.

Technology Category

• Machine Learning: Hardware-aware ML
• Planning, Routing, and Scheduling: Model-Based Reasoning
• Humans and AI: Human-in-the-loop Machine Learning

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• Responsible Web: Machine-in-the-loop, human agency and autonomy
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs

Abstract

Wireless communications are characterized by their unpredictability, posing challenges for maintaining consistent communication quality. This paper presents a comprehensive analysis of various prediction models, with a focus on achieving accurate and efficient Wi-Fi link quality forecasts using machine learning techniques. Specifically, the paper evaluates the performance of data-driven models based on the linear combination of exponential moving averages, which are designed for low-complexity implementations and are then suitable for hardware platforms with limited processing resources. Accuracy of the proposed approaches was assessed using experimental data from a real-world Wi-Fi testbed, considering both channel-dependent and channel-independent training data. Remarkably, channel-independent models, which allow for generalized training by equipment manufacturers, demonstrated competitive performance. Overall, this study provides insights into the practical deployment of machine learning-based prediction models for enhancing Wi-Fi dependability in industrial environments.

Problem

• Predicting Wi-Fi link quality accurately using machine learning techniques
• Developing low-complexity models suitable for resource-limited hardware platforms
• Enhancing Wi-Fi dependability in industrial environments through practical deployment

Innovation

• Machine learning models for Wi-Fi link prediction
• Low-complexity exponential moving average combination
• Channel-independent training for generalized deployment