IoT-MCP: Bridging LLMs and IoT Systems Through Model Context Protocol

2025-09-25

AI Summary

To address the challenges of high hardware heterogeneity and complex control logic in LLM–IoT integration, this paper proposes the first edge-deployment framework tailored for LLM–IoT co-execution. Methodologically, we design a lightweight Model Context Protocol (MCP) and an accompanying edge-server architecture to establish a unified communication protocol stack, supporting 22 sensor types and 6 MCU families. We further introduce IoT-MCP Bench—the first standardized benchmark for LLM–IoT systems—comprising 114 primitive tasks and 1,140 composite tasks. Our contributions include open-sourcing a full-stack framework enabling real-time tool invocation and context-aware responses. Experimental results demonstrate 100% task success rate, an average response latency of 205 ms, and peak memory consumption of only 74 KB—significantly enhancing accuracy, efficiency, and reproducibility of LLM-driven physical device control.

Technology Category

• Machine Learning: Learning on the Edge & Model Compression
• Application Domains: Internet of Things, Sensor Networks & Smart Cities
• Intelligent Robots: Multimodal Perception & Sensor Fusion

Application Category

• Economics, Online Markets and Human Computation: Cost models of using LLMs in production systems
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions

Abstract

The integration of Large Language Models (LLMs) with Internet-of-Things (IoT) systems faces significant challenges in hardware heterogeneity and control complexity. The Model Context Protocol (MCP) emerges as a critical enabler, providing standardized communication between LLMs and physical devices. We propose IoT-MCP, a novel framework that implements MCP through edge-deployed servers to bridge LLMs and IoT ecosystems. To support rigorous evaluation, we introduce IoT-MCP Bench, the first benchmark containing 114 Basic Tasks (e.g., ``What is the current temperature?'') and 1,140 Complex Tasks (e.g., ``I feel so hot, do you have any ideas?'') for IoT-enabled LLMs. Experimental validation across 22 sensor types and 6 microcontroller units demonstrates IoT-MCP's 100% task success rate to generate tool calls that fully meet expectations and obtain completely accurate results, 205ms average response time, and 74KB peak memory footprint. This work delivers both an open-source integration framework (https://github.com/Duke-CEI-Center/IoT-MCP-Servers) and a standardized evaluation methodology for LLM-IoT systems.

Problem

• Bridging LLMs and IoT systems through standardized communication protocol
• Addressing hardware heterogeneity and control complexity in LLM-IoT integration
• Providing framework for IoT-enabled LLM task execution and evaluation

Innovation

• IoT-MCP implements Model Context Protocol for LLM-IoT integration
• Edge-deployed servers bridge language models with physical devices
• Framework achieves 100% task success with 205ms response time