IRS Empowered UAV Wireless Communication with Resource Allocation, Reflecting Design and Trajectory Optimization
Xiaoqi Zhang,Haijun Zhang,Wenbo Du,Keping Long,Arumugam Nallanathan
DOI: https://doi.org/10.1109/twc.2022.3162704
IF: 10.4
2022-01-01
IEEE Transactions on Wireless Communications
Abstract:As revolutionary technologies that can actively change the communication link signal, intelligent reflecting surface (IRS) and unmanned aerial vehicle (UAV) have emerged as reliable, economical and convenient wireless communication solutions for a variety of practical scenarios. Therefore, this paper focuses on an IRS empowered UAV downlink communication network, where the dynamic UAV establishes a cascade link via IRS to provide signal enhancement services for multiple users. Considering constraints of transmit power, flight speed and area at the UAV and the reflecting constraints at the IRS, the block coordinate descent (BCD) method based on resource allocation, reflecting design and trajectory optimization is adopted to maximize the sum-rate of all users. The proposed problem is converted by using quadratic transformation and Lagrangian dual transformation. Then applying for the approximate linear method and Iterative Rank Minimization (IRM) to optimize the transmit power of UAV and phase shift of IRS respectively. Since additional reflection propagation paths by IRS, the complexity of the channel model makes the trajectory design difficult. To tackle this problem, this paper proposes a UAV trajectory optimization method based on enhanced reinforcement learning with the fixed initial location and destination. In the end, the convergence of the proposed scheme is effectively verified by simulations. Moreover, abundant simulation comparisons between the proposed scheme and other benchmark schemes demonstrate the validity and high performance gains of the proposed algorithm.
telecommunications,engineering, electrical & electronic