A method for IoT devices test case generation using language models

The rapid growth of IoT and electronic systems has led to complex real-time data processing and management solutions. However, these systems present significant software and hardware testing challenges, often requiring manual, time-consuming testing efforts. To address this, an automated end-to-end...

Full description

Saved in:
Bibliographic Details
Main Authors: Sumit Kumar, Kiran Napte, Ruchi Rani, Sanjeev Kumar Pippal
Format: Article
Language:English
Published: Elsevier 2025-06-01
Series:MethodsX
Subjects:
Online Access:http://www.sciencedirect.com/science/article/pii/S2215016125001864
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:The rapid growth of IoT and electronic systems has led to complex real-time data processing and management solutions. However, these systems present significant software and hardware testing challenges, often requiring manual, time-consuming testing efforts. To address this, an automated end-to-end testing framework is essential for improving efficiency and reliability in IoT system development. With advancements in Natural Language Processing (NLP) and language models, automated test case generation systems can now create structured test cases in programming languages while ensuring code integrity and style. Applying these techniques to IoT projects streamlines testing, enhances accuracy, and reduces workload. This paper introduces TCG-IoT (Test Case Generation for IoT Systems), an automated testing framework designed to generate comprehensive test cases and actionable event lists based on technical and data specifications. Unlike existing frameworks that depend on model-based testing or simulation environments, TCG-IoT uniquely integrates a Retrieval-Augmented Generation (RAG) mechanism, a vector knowledge base of IoT standards, and code generation via Code-Llama to directly produce structured, executable C-code scripts for software and manual steps for hardware components. The results demonstrate that TCG-IoT delivers high-quality, context-aware test cases and scripts with maximum system coverage, ensuring secure, efficient, and scalable IoT development. • Automates test case generation for IoT hardware and software components. • Enhances test coverage and reliability using language models. • Evaluates performance through case studies on smart home automation.
ISSN:2215-0161