Identification of connected objects based on network protocol data

The increasing number of connected objects and their growing variety make it necessary to have a system to recommend IoT services for end-users, so that they can make better use of the connected objects at their disposal. In this context, we would like to propose autonomous mechanisms communicating with the recommendation system and aiming to facilitate the management of services and objects. In this article, we deal in particular with the problem of identifying connected objects in a home network. For this, we propose different approaches to solve this problem. The first type of approach is based on network protocols such as discovery protocols. The results show that thanks to the extracted information such as the model of the object, the vendor, the operating system, we manage to identify the type of the object. A second approach is based on traffic analysis using machine learning techniques that will be addressed in future work.