| （英） |
Aerial base stations (ABSs), which are cellular base stations mounted on unmanned aerial vehicles (UAVs), have gained attention for expanding coverage and capacity in future wireless networks. Owing to their flexible mobility, ABSs can adaptively respond to a sudden traffic increase, and by visiting ground users, they can efficiently provide communication environments to widely distributed users. However, designing the optimal trajectory that maximizes the communication quality of users for multiple ABSs is challenging due to various factors, such as the high degree of freedom of the problem and inter-cell interference. For this problem, various trajectory optimization methods have been proposed; however, most of them are offline and centralized approaches in which the future channel status and the positions of all users and UAVs are assumed to be known. Thus, they are not robust to the randomness of channels, and the computational costs become large as the area and number of users increase. In this study, we propose an online and distributed trajectory control method based on distributed model predictive control for multiple ABSs. We consider a scenario in which multiple ABSs provide downlink communication to ground users while moving towards their destinations. We formulate a joint optimization problem of ABS trajectory and user association that minimizes the flight time and power consumption of UAVs while satisfying the communication quality of users. In the proposed method, each ABS constructs its trajectory sequentially based on model predictive control in a distributed manner using only local information. Furthermore, we evaluate the proposed method by simulation and demonstrate its effectiveness. |