A Modular, Low-Cost IoT System for Environmental and Behavioural Monitoring in Cultural Heritage Sites

The preservation of cultural heritage faces growing challenges from climate change, tourism pressure, and limited conservation resources. Existing monitoring solutions are often cost-prohibitive, proprietary, and inflexible, leaving many institutions, particularly in developing regions, without viable tools for proactive management. This study presents a modular, low-cost Internet of Things (IoT) system designed for real-time environmental and behavioural monitoring in heritage sites. Built with off-the-shelf components such as ESP32 microcontrollers and Raspberry Pi, the system integrates a wireless sensor network, edge computing, and cloud services (Microsoft Azure) to measure variables including temperature, humidity, sound, and visitor proximity. It also incorporates computer vision models to classify visitor behaviour, achieving up to 95% accuracy using fine-tuned Vision Transformers. The system's modularity, enabled via JSON configurations, allows for rapid reconfiguration without firmware changes. A simulated deployment demonstrated robust performance, low power consumption, and cost-efficiency (less than 200 GBP per node), validating the system's potential for scalable, sustainable heritage monitoring. This open-source framework offers a practical path forward for institutions seeking to balance accessibility with conservation needs.