The combination of free space optic (FSO) and unmanned aerial vehicle (UAV) can be a promising solution to last-mile problem because of FSO high bandwidth and flexibility of UAV. However, FSO is vulnerable to the atmospheric turbulence and transmitter configuration is limited on UAV. Therefore, in this paper, we investigate an adaptive resource allocation strategy to provide relative fair and high transmission capacity for users. In the proposed network model, the downlink scenario is considered and ground station communicates with UAV and its local users by FSO and RF links, respectively. According to the channel conditions and rate requirements from users, channel and power assignments should satisfy the capacity demand in reasonable. To this end, we decompose the original channel and power allocation problem with high computational complexity into low-complexity subproblems, corresponding to channel allocation in RF transmission phase and channel matching and power allocation in FSO transmission phase. Then, a heuristic efficient resource assignment algorithm is proposed to achieve the optimal capacity distribution for users. Numerical results show that the proposed method can asymptotically achieve optimal throughput. Furthermore, system throughput is higher at low transmitting power requirement than those of other existing methods.

自由空间光学系统（FSO）和无人机（UAV）的结合可能是解决最后一英里问题的有前途的解决方案，因为FSO具有高带宽和无人机的灵活性。但是，FSO易受大气湍流的影响，并且无人机上的发射机配置受到限制。因此，在本文中，我们研究了一种自适应资源分配策略，以为用户提供相对公平和较高的传输容量。在提出的网络模型中，考虑了下行链路情况，地面站分别通过FSO和RF链路与UAV及其本地用户进行通信。根据用户的信道条件和速率要求，信道和功率分配应合理满足容量需求。为此，我们将具有较高计算复杂度的原始信道和功率分配问题分解为低复杂度子问题，对应于RF传输阶段的信道分配以及FSO传输阶段的信道匹配和功率分配。然后，提出一种启发式高效资源分配算法，以实现用户的最佳容量分配。数值结果表明，该方法可以渐近地实现最优吞吐量。此外，在低发射功率要求下，系统吞吐量比其他现有方法更高。提出了一种启发式高效资源分配算法，以实现用户的最佳容量分配。数值结果表明，该方法可以渐近地实现最优吞吐量。此外，在低发射功率要求下，系统吞吐量比其他现有方法更高。提出了一种启发式高效资源分配算法，以实现用户的最佳容量分配。数值结果表明，该方法可以渐近地实现最优吞吐量。此外，在低发射功率要求下，系统吞吐量比其他现有方法更高。