Abstract
With the development of unmanned aerial vehicle (UAV) technology, using UAVs as aerial communication platforms to provide additional communication services is a promising technology. However, due to the scarcity of spectrum resources, UAV will inevitably causes unintentional interference to primary users (PUs) when transmitting data to secondary users (SUs). Aiming at this problem, with the help of the multi-objective optimization problem (MOOP) framework, this paper studies the trade-off between UAV's communication rate and UAV's unintentional interference by optimizing UAV's trajectory, UAV's transmission power, and SU's scheduling. Specifically, we consider a cognitive UAV communication scenario that a single UAV needs to transmit data to multiple SUs in a shared spectrum scenario. During this process, if the existing PUs receive unintentional interference from the UAV does not exceed a given threshold which is also called interference temperature (IT), it will be regarded as a normal communication state. Then, a MOOP is established that includes maximizing the minimum UAV-SU communication rate and maximizing the times that PUs are in normal communication states. Finally, with the help of \epsilon -constraint method and stochastic submodular optimization algorithm, we propose an iterative algorithm to obtain the approximate Pareto front. The tradeoff between UAV's communication rate and UAV's unintentional interference is analyzed through numerical simulation.