ABSTRACT
In recent years, IoT devices have become widespread in households, and IoT devices with various functions such as remote control, lighting, door locks, and power outlets are sold and used in various situations. The activity of IoT devices can be likened to a black box, often operating independently of the user's intentions regarding what data the device is sending and where. Therefore, we are aiming to realize a framework called the IoT activity tracker that has a function of access control, which can detect what kind of communication IoT devices are doing and allow only appropriate communication based on it, and a function that enables users to understand the operation status of IoT devices by visualizing what kind of communication IoT devices are doing. In order to achieve the IoT activity tracker, in this paper, we collected communication traffic of 16 combinations of device model and executed function from 8 IoT devices with 2 functions each. And we used 28 features derived from the volume of communication that do not contain information that can identify individuals or specific manufacturers to perform 2 types of classification using a random forest algorithm and evaluated their accuracy. As a result, we confirmed that the function could be estimated with an accuracy of 91% when classified into 16 combinations of device model and executed function. When classified by 8 combinations of only executed functions, we confirmed that the function could be estimated with an accuracy of 73%.
