Resource Allocation And Optimization Of UAV-aided Multi-user Full-duplex Wireless-powered Communication Networks

Based on the energy supply difficulty of low-power sensor nodes in next-generation wireless communication scenarios(such as the Internet of Things,wireless sensor networks),and the fixed hybrid access point technology in traditional wireless-powered communication networks(WPCN)is no longer applicable to future complex scenarios with multiple users.For the current new type of unmanned aerial vehicle(UAV)technology,which has the characteristics of high mobility,wide range,can be equipped with other equipment,strong endurance,and so on.This thesis proposes a new UAV-aided full-duplex WPCN system,in which the UAV can flexibly adjust its position.It ensures a good communication channel with ground users,and also brings more challenging solutions for UAV deployment and resource allocation.Thus,it is of great and practical significance to explore the resource allocation and optimal design of the new system in various scenarios.Therefore,according to the degree of distribution of users,this thesis mainly focuses on the following two scenarios:1)For sparse user distribution scenarioFirst,a one-dimensional UAV-aided multi-user full-duplex WPCN model is established,where a UAV performs downlink(DL)energy transmission to ground users in turn.At the same time,users transmit uplink(UL)information in a time division multiple access(TDMA)mode to UAV.And two user energy harvesting(EH)schemes are designed based on the characteristics of the UAV: i)hover-only EH scheme,where the UAV can only charge users when hovering;ii)hover-and-fly EH scheme,where the UAV can charge users when hovering and flying.Further,under the two EH schemes,the optimal time resource allocations for maximizing network throughput and minimizing the total task time of the UAV are studied,respectively.The new system has the following characteristics: i)The energy causality is considered into our system,i.e.,each user can only use all the energy harvested so far before its transmission.ii)The flying time of the UAV and the maximum speed constraint of the UAV are considered into the system.iii)The wireless power transmission range of the UAV is limited in the system due to the actual receiver requirements.The final results show that the relationship between the hovering time and the flying time obtained under the two optimization goals is exactly the opposite.Also,the hovering-and-fly EH scheme outperforms the hover-only EH scheme and traditional fixed base station(BS)schemes.As the flying height of the UAV decreases,and the distance between the sensors decreases or the number of sensors increases,the sum throughput will gradually increase and then stabilize.2)For dense user distribution scenarioFirst,a three-dimensional(3D)UAV-aided multi-user full-duplex WPCN model is established,where a swarm of UAVs is deployed in 3D simultaneously charge all ground users and then fly to new locations to collect information from scheduled users during several epochs via orthogonal frequency division multiple access(OFDMA).Due to potential limited channels issue in Narrow Band Internet of Things,two scenarios are divided: the number of channels is greater than the number of users,and the number of channels is less than the number of users.In these two scenarios,an optimization framework is proposed to jointly optimize the UL-and-DL 3D deployment of the UAV swarm,including the user-UAV association,the scheduling order,and the UL-DL time allocation.Under the proposed framework,several effective algorithms are summarized to solve the problem.The new system has the following characteristics: i)The DL EH threshold for users and the UL signal-to-noise ratio threshold for UAVs are taken into consideration when designing the system ii)a new channel model considering both line-of-sight and non-line-of-sight channels of multiple altitude-adjustable UAVs is applied.Besides,according to the characteristics of the system,two suboptimal scheduling strategies are designed,referred to as the near-first(NF)scheme and far-first(FF)scheme.The final results show that the 3D position distribution of the UL and DL UAVs has different trends,and the location distribution of the DL UAVs has a convergence phenomenon;the performance of the proposed model outperforms that of the traditional fixed BS WPCN;the OFDMA scheme significantly outperforms the TDMA scheme;In terms of maximizing throughput,the NF strategy outperforms the FF strategy,and in terms of system fairness,the FF strategy outperforms the NF strategy.