Enhancing IoT-Driven Smart Home Security and Automation with a GCN Model

A network of computing devices and sensors, the Internet of Things (IoT) rapidly exchanges data to assess circumstances and propose new services. Intelligent home technologies rely on IoT. Contemporary smart home systems offer automatic lighting control, smoke detectors, temperature sensors, and smart lock installations. But these technologies raise serious privacy and security concerns. Security flaws let hackers access sensitive data, sendoff false alarms, and take control monitoring systems. Smart home technology is vulnerable to many threats because consumers are wary of it. Recommended steps are preprocessing, sensing, and model training. Preparing raw, unstructured data for analysis. DHT11 humidity and temperature sensors and HC-SR501 PIR motion detectors collect data. It trained the model using Multiscale DGCN data. Better than GCN and BiLSTM, Multiscale DGCN performs. It uses Internet of Things-based smart home security data well, with an average accuracy rate of 93.87%. Identifying outliers in sensor data solves major smart home security challenges. Increasing smart home system accuracy reduces security concerns and boosts user trust, which may drive adoption.