Enabling Cloud-Level Accuracy in Edge AI through IoT Data Preprocessing

2026-06-21

AI Summary

This work addresses the challenge of deploying large language models (LLMs) on edge devices for processing raw IoT sensor data, where limited numerical reasoning capabilities often compromise both accuracy and latency. The authors propose a lightweight, structured prompt construction framework that transforms raw sensor readings—such as those from a BME680—into progressively enriched textual representations, including raw values, threshold-aware descriptions, and environmental summary tags. By integrating chain-of-thought (CoT) reasoning at the input stage without modifying the underlying model architecture, the approach substantially narrows the performance gap between edge and cloud-based LLMs. Experimental results demonstrate significant improvements: indoor task accuracy rises from 50.9% to 81.7%, and outdoor accuracy from 63.7% to 89.3%. Moreover, in CoT-free mode, average inference latency remains as low as 0.22 seconds, enabling high-accuracy, low-latency edge intelligence.

Abstract

Large language models (LLMs) offer a natural-language interface for interpreting Internet of Things (IoT) sensor data in smart environments; however, cloud deployment introduces latency, privacy, and connectivity concerns. Local LLMs can reduce these limitations, but compact edge-deployable models often show weaker numerical reasoning when raw sensor readings are provided directly. This paper investigates whether prompt-side preprocessing can improve the accuracy-latency trade-off of local LLMs for environmental monitoring. We propose a structured prompt construction framework that transforms raw air-quality and thermal-comfort measurements into progressively enriched textual representations: raw sensor values, threshold-aware descriptions, and compact environmental summary flags. The approach is evaluated using indoor Raspberry Pi/BME680 datasets from Tampere University and outdoor air-quality datasets from Helsinki, Katowice, and Warsaw. We construct a binary LLM query dataset covering air quality, thermal comfort, and joint environmental conditions, and evaluate five local and five cloud LLMs across three prompt variants and two inference modes, with and without chain-of-thought prompting. Results show that prompt enrichment substantially improves local-model accuracy. In No-CoT mode, local accuracy increases from 50.9% to 81.7% indoors and from 63.7% to 89.3% outdoors from the raw to the most enriched prompt. Local No-CoT inference is the fastest configuration, with mean latency close to 0.22 s, while CoT substantially increases inference time. These findings suggest that lightweight prompt-side preprocessing can narrow the local--cloud performance gap and support low-latency IoT analytics in smart environments.

Problem

• Edge AI
• IoT data preprocessing
• Large Language Models
• Accuracy-latency trade-off
• Environmental monitoring

Innovation

• prompt preprocessing
• edge AI
• IoT data
• local LLMs
• accuracy-latency trade-off