Distributed Services Management over Dynamic Networks of Things: towards a Unified SDN oriented Design

The work presented in this dissertation shows another possible SDN adoption in the WSANs and more in general in the IoT ecosystem, rethinking the integration logic of the two paradigms by surpassing the limitations of the existing approaches. The direct management of the SDN Controller has been used to dynamically change the routing policy of the WSAN by minimizing the energy consumption of the sensor nodes and optimizing the use of underutilized network resources. Focusing on environments where devices are battery powered and installed in harsh and hardly accessible locations, the Wireless Power Transfer (WPT) technique has been considered to improve the overall networks lifetime. The resulting SD-WSAN architecture has been enriched by a new virtual level, called Power Plane and the recharging procedure has been combined with the management of the SDN Controller to maximise the services availability and system durability of clustered wireless IoT domains. In this scenario, the benefi ts of advanced energy-saved techniques and smart network strategies are often limited by Web of Things (WoT) communication protocols which are not optimized for wireless sensor networks (e.g. Message Queue Telemetry Transport (MQTT)). In order to overcome this open issue, the proposed model has been used to modify the default behaviour of existing application protocols and enable innovative traffic flows. Finally, the study moved from static to dynamic environments and the Internet of Vehicles (IoV) sub-domain has been analysed. IoV allows vehicles to share information about their surroundings, traffic condition, and other relevant data, enabling them to make more informed decisions and improving safety, efficiency and convenience. However, traditional Cloud approaches introduce latencies that make the development of innovative real-time applications infeasible. To this purpose, SDN has been applied in combination with the Fog Computing (FC) paradigm to properly accommodate users mobility and to support complex mobile services with stringent time-space constraints.