WiFi Pathologies Detection using LLMs

TL;DR

This paper explores the use of fine-tuned large language models to detect WiFi network issues, demonstrating high accuracy with both labeled and unlabeled data.

Contribution

It introduces a novel application of LLMs for WiFi pathology detection, comparing encoder-only and decoder-only models in this context.

Findings

High detection accuracy with labeled data

Decoder-only models perform well with unlabeled data

Sequential models outperform in certain scenarios

Abstract

In this paper, we fine-tune encoder-only and decoder-only large language models (LLMs) to detect pathologies in IEEE 802.11 networks, commonly known as WiFi. Our approach involves manually crafting prompts followed by fine-tuning. Evaluations show that the sequential model achieves high detection accuracy using labeled data, while the causal model performs equally well for unlabeled data.