Resource Allocation for UAV-Assisted IoT Networks with Energy Harvesting and Computation Offloading

This paper considers an unmanned aerial vehicle (UAV) assisted internet of things (IoT) network. In order to prolong the lifetime of power-limited IoT devices (IDs), it is assumed that UAVs can act as energy sources and provide radio frequency (RF) energy for IDs. Moreover, to relieve the tension between energy-limited IDs and computation-intensive applications, assume that the tasks of IDs can be offloaded to UAVs. Based on these settings, we aim to minimize the energy consumption of UAVs. Due to the integer feature of the ID association matrix, the formulated problem is a mixed integer program, which is usually difficult to solve. To handle this problem, the original problem is decomposed into three subproblems, and it is shown that each subproblem can be optimally solved. An alternative algorithm is then provided. Simulation results show that the energy consumption of UAVs can be effectively decreased by the proposed algorithm, and compared with the non-offloading scheme, the total energy consumption can be significantly reduced by computation offloading.