Real-Time IoT Device Identification Using Traffic Pattern Mapping Model

With the widespread adoption of smart homes, automatic identification of IoT devices and their behavior is becoming increasingly important for maintaining safe, secure, and comfortable smart home environments. To grasp and understand IoT devices and their behavior, various IoT device identification methods based on traffic analysis with machine learning have been proposed and realized to automatically classify IoT devices with high accuracy. We also proposed an IoT traffic analysis and device identification method based on two-stage clustering previously. However, it has not considered real-time processes in actual environments. This paper proposes a real-time IoT device identification method using a traffic pattern mapping model for smart home environments. By using the analytical results as the mapping model to assign IoT traffic patterns, which can be obtained by traffic analysis with the clustering method, we can successfully reduce the computational cost required for pattern assignment and adapt our previous method to real-time IoT device identification.