Data traffic analysis to monitor and understand the people's mobility in Smart Cities

The Smart City is an abstract projection of future communities, a virtual fence defined by a set of needs that find answers in technologies, services and applications which can be traced back to different domains: smart building, inclusion, energy, environment, government, living, mobility, education, health, and much more. In general, the smart city concept involves different ICT technologies and several physical devices which are connected to the Internet of Things to optimize the efficiency of city services and with the objective to improve the quality of life of the citizens. In this context the most strenuous and crucial activity is to collect and organize useful data to plan and improve city’s services. In this thesis I will focus on the may cheapest way to collect a high spatial granular human mobility data: Wi-Fi data traffic. The traffic is analyzed in a passive and anonymous way, the privacy of the users is totally preserved. In the piratical side the Wi-Fi traffic which is relevant for this work is composed by Probe Request frames. Those frames are the packets sent by Wi-Fi enabled devices during the phase of Access Point discovering, in the context of IEEE 802.11 protocol, such phase is called Active Scanning. The main contribute of my PhD journey was in the design and developing of complete cloud-enabled IoT system and in the design of a MAC address de-randomization algorithm based on the combination of temporal- and content-basedprobe request fingerprints. The developed algorithm is Artificial Intelligent based and could be used to count and track Wi-Fi devices in an anonymous way.