Smart devices fingerprint detection

The generalized use of mobile devices with Internet connectivity for both 3G and WiFi, allow users to choose the connection they want to use at all times. This behavior requires that Internet service providers must adapt their infrastructure to ensure good levels of Quality of Service in both types of connections. In this paper, we describe an intelligent system based on neural networks and finite state machines that lets the Internet service provider know to which type of device belongs the traffic going to its network. The system analyzes the transport and application layers from TCP packets to discriminate the percentage of Internet traffic generated by mobile devices and personal computers. Test results show the success of the developed system.