UAV-based Data Collection For Wireless Sensor Networks

With the vision of 6G to establish an integrated space-air-ground-underwater network,and the advantages provided by the UAV,which enables high mobility,rapid and flexible deployment,and line-of-sight(Lo S)dominant channel,the UAV-assisted wireless communication has been widely applied in consumer electronics,vertical industries,emergency communication,etc.It will be envisioned that the applications from the vertical industries,specifically in the wireless sensor networks(WSNs)including internet of things(Io T),industrial Io T(IIo T),will further promote the advance of the UAV-aided communications.The WSN lifetime and performance depend on the energy available at the sensor nodes.The traditional WSN has a fusion center to collect and process the data from the sensor node via multi-hop transmissions,thereby accelerating the network energy consumption.Besides,it is difficult to deploy a WSN in some dreadful conditions,while the network connectivity and transmission performance cannot be guaranteed.Therefore,this thesis considers the deployment of UAV for WSN data collection.However,the onboard energy at the UAV will also limit the performance of the WSN.And the high mobility of the UAV leads to the dynamic change of the UAV-ground channels.So it is essential to carefully optimize the UAV trajectory for more efficient data transmission.To this end,this thesis optimizes the UAV-assisted data collection schemes by taking the energy of both sensors and UAVs into account.The main contributions of this thesis are as follows.This thesis first considers a UAV to collect data from sensors deployed in an area of interest and studies two orthogonal multiple access schemes for ground sensors,i.e.time division multiple access(TDMA)and frequency division multiple access(FDMA),respectively.On the premise of meeting constraints of the energy and the amount of data transmitted for sensors,the UAV trajectory,the sensor's transmission power and resource allocation(time or bandwidth)are jointly optimized to minimize mission completion time.A bisection method with each step employing the successive convex approximation(SCA)is proposed to solve the non-convex mixed integer programming problem and the consistency of TDMA and FDMA schemes is revealed.The performance of the proposed algorithm is verified and evaluated by numerical simulations.It shows the UAVs are prone to approach the sensor nodes if the amount of data increases or the available energy decreases.This thesis further extends the design to a UAV swarm scenario to achieve more efficient data collection.All UAVs will be allocated with an orthogonal frequency band,and the ground sensors will adaptively be connected to at most one UAV at each time slot by the TDMA scheme.Two cases,i.e.,fixed and adaptive bandwidth allocation of UAVs are investigated.By optimizing the flight trajectory and bandwidth allocation of each UAV,as well as the transmission power,time allocation and UAV association of each sensor,the mission completion time is minimized.This thesis proposes a bisection method and an alternating optimization algorithm with SCA and non-convex penalty to resolve the challenges due to the integer and sparsity constraints.Moreover,based on the results of adaptive bandwidth,the UAVs' trajectories and bandwidth allocation and sensors' energy and resource allocation,etc.,are further optimized to minimize the energy consumption of the sensors.This thesis also validates the performance of the algorithms by simulations and compares the performance for the considered schemes.It shows that proposed bandwidth allocation scheme introduces great flexibility to the system design.