Minimization of Transmission Completion Time in UAV-Enabled Wireless Powered Communication Networks

Authors:
Publisher website
Google Scholar
Abstract
This paper considers the unmanned aerial vehicle enabled wireless powered communication networks (UAV-enabled WPCN) where one UAV plays the role of hybrid sink (H-sink) and coordinates the wireless energy/information transmissions to/from a set of nodes, and aims to minimize the transmission completion time (TCT) of collecting a given number of bits per node. Due to its intractability, this paper transforms this problem into a tractable one by using an area discretization technique so that the node’s energy harvesting power can be approximated to be the same value under a given error tolerance ε wherever the UAV is located inside one subregion. The optimal solution of transformed problem has an approximation ratio of 1+ε to the theoretically minimum TCT. To solve the transformed problem, this paper formulates it as a convex problem and decomposes it into the master problem and the slave linear programming problem. The master problem is solved by a subgradient based algorithm. Furthermore, for the scenario where each node has the same amount of data to transmit, this paper develops an algorithm with lower complexity. Specifically, this paper first decomposes it into the master problem of determining the minimum TCT via the bisection search method and the slave feasibility problem under a given TCT. The slave problem is transformed to be a convex problem whose optimal solution is obtained by partially solving its Lagrange dual problem first and then solving a linear programming problem. Simulation results demonstrate that the UAV-enabled WPCN greatly outperforms the conventional WPCN with the fixed H-sink.
Funding Information
  • National Natural Science Foundation of China (61872322, 61902353)
  • Zhejiang Natural Science Fund for Distinguished Young Scholars (LR20F020003)
  • Natural Science Foundation of Zhejiang Province (LY18F020029)

This publication has 43 references indexed in Scilit: