Recent advances in Unmanned Aerial Vehicle (UAV) technologies make their use as data collection (sink) feasible for practical applications of Wireless Sensor Networks (WSN) and the Internet of Things (IoT). Using UAVs as mobile data ferries can amplify ground network performance by mitigating the impact of horizontal radio pollution on the environment and replacing that with individual nodes or local group of nodes vertically communicate with opportunistically accessible UAV. This can reduce packet loss performance and energy use on a per device basis, as reduced routing responsibilities require significantly less radio activity.

This research aims to develop a UAV-aided WSN model that provides energy efficient and scalable data gathering method and supports numerous applications such as forest fire monitoring, monitoring of microclimates, flood monitoring and monitoring of forest predator’s traps, to name few possible applications. One of the challenges for the UAV-assisted WSN for data gathering efforts is the design of an energy-efficient UAV communication with randomly distributed ground sensors by improving the ground network structure dependent on the UAV path. Hence, in keeping with the context of UAV energy efficiency challenges and solutions, the proposed model incorporates three parts; the UAV trajectory planning in which the smooth UAV route is constructed, the UAV-Ground connectivity wherein the UAV data link communication is initiated based on the UAV’s position in the selected path, and the wireless communication within the ground sensor nodes (SN) wherein sensors collaborate to organise data transmission to the UAV.

The first part is focused on identifying a flexible span for the UAV path design by conceptualizing an approach called ‘UAV Fuzzy Travel Path’ that supports UAV smooth trajectory planning and facilitates ground network re-orchestration for data gathering from spatially dispersed wireless sensors over a large space. This organisation enables the UAV path to be dynamically adjusted in accordance with the updated ground network structure and allows for defining an adaptable span for UAV path design instead of being fixed into a static route. Crisp circular/linear UAV path and Bezier Curve TSP (BTSP) path are two examples of smooth paths aligned on a variety of ground network formations here.

The second part proposes the dynamic orchestration of ground wireless sensors grouping for improving the network performance in data collection. The term 'Software Defined Wireless Sensor Network' (SDWSN) will be used broadly in this thesis to describe the flexibility of the ground network architecture. This offers a more dynamic and adaptive network that interacts with the UAV planned path. The SDWSN-enabled ground network communication is mainly divided into two major phases: the scanning topological pre-orchestration and sensing data collection post-orchestration phase. As the primary goal of data collection effort is on passing the sensing data information messages to the drone via data collection phase, passing the ground network topological orchestration data and assigning the optimal functionalities to each component of the network are operated via scanning pre-orchestration phase.

The third part presents the key prerequisites for designing a generic model for air-to-Ground network that takes the mobile network topology into account. The air-to-Ground communication structure via the proposed communication window of connectivity mechanism is expressed to provide a connection among the ground network and the UAV. The proposed communication design is identified based on two approaches: air-to-Ground communication with each individual sensor nodes or air-to-Ground communication with each group representative nodes serving as cluster heads. Both techniques attempt to offer a reliable and consistent communication among the sensor network and the drone.

The main evaluation and performance metrics for the first part focus on testing the UAV propulsion energy consumption and received packets on the UAV receiver while accounting for smooth paths. The theoretical and actual analysis of the proposed UAV path designs prove that the designed crisp circular/linear UAV path improves energy efficiency for propulsion energy consumption with higher packet delivery rate on the UAV receiver while compromising ground network energy usage compared to the BTSP. The key performance measures for the analysis of ground network are packet delivery and energy consumed for the overall ground network when dealing with different network densities and message rates. The analytical results of both scanning and sensing data collection phases mentioned above confirm that the ground network orchestration plays a crucial role in serving more SNs on the ground and enhancing UAV energy efficiency when compared to the situation prior to orchestration. Additionally, the impact of UAV velocity and network distribution on the percentage of received packets on the UAV receiver is evaluated in the air-to-Ground communication evaluation section. The simulation outputs show that the proposed SDWSN-enabled data gathering modelling boosts packet delivery rate on the UAV-Ground connectivity when compared to the situation prior to network orchestration.