Simulating Devices in a Smart Home Environment

Testing IoT devices can be quite challenging, primarily due to the need for real hardware to effectively simulate how these devices interact in live environments, such as Smart Home. Unlike software, which can often be tested in virtual settings, IoT devices require physical setups to account for network conditions, interoperability, and real-world response times. This reliance on physical devices, raising costs and reducing scalability, makes it difficult to test complex or large-scale IoT networks comprehensively. This thesis aims to solve this problem, by designing and developing a simulator of Commercial-Off-The-Shelf smart devices, in this case Shelly devices, replicating their functionalities and their interactions among one another and within the Smartotum ecosystem and the Home Assistant platform. The project’s main objective is to create a cost-effective and flexible testing environment for smart home solutions without the need for physical devices. Using Object-Oriented Programming (OOP) principles in Python, the simulator models a variety of Shelly devices, limited to the first generation, and integrates MQTT and REST communication protocols to mimic real-world smart device interactions. The devices also integrate with Smartotum, a decentralized smart home solution that emphasizes privacy, interoperability, and reliability by eliminating the reliance on cloud-based operations, relying instead on a distributed hash table. The Smartotum ecosystem overcomes typical vendor lock-in issues common in the smart home industry, allowing different devices to operate together in a unified environment. The simulator design structure enables testing not only of single-device interactions but also of multi-device scenarios where numerous devices communicate and interact in parallel, simulating a cohesive smart home ecosystem. Functionality tests were conducted to assess Smartotum’s handling of real-world automation, specifically through the simulation of a heating system controlled by temperature sensors. Mocked temperature curves were used to emulate daily temperature fluctuations, and the heating system was programmed to respond dynamically by adjusting its output based on these curves. Through rigorous testing, the simulator demonstrates the effectiveness of the Smartotum ecosystem in real-time smart home interactions. These tests highlight the simulator’s value as a robust platform for evaluating smart home automation and facilitating future enhancements across smart home technology. The results emphasize its potential as a scalable, cross-platform tool, offering significant advantages for developers and researchers working to improve smart home capabilities.