User Association and Bandwidth Allocation for Terrestrial and Aerial Base Stations with Backhaul Considerations

TL;DR

This paper proposes a joint optimization framework for user association, bandwidth allocation, and 3D placement of drone base stations in a hybrid terrestrial-aerial network, considering backhaul constraints and ultra-reliable low-latency communication requirements.

Contribution

It introduces a novel algorithm combining decomposition and heuristic optimization to efficiently position DBSs and allocate resources in complex network scenarios.

Findings

The proposed method improves user rate fairness and network throughput.

Backhaul link quality significantly impacts DBS placement and performance.

Traffic distribution and antenna beamwidth influence optimal deployment strategies.

Abstract

Drone base stations (DBSs) can enhance network coverage and area capacity by moving supply towards demand when required. This degree of freedom could be especially useful for future applications with extreme demands, such as ultra reliable and low latency communications (uRLLC). However, deployment of DBSs can face several challenges. One issue is finding the 3D placement of such BSs to satisfy dynamic requirements of the system. Second, the availability of reliable wireless backhaul links and the related resource allocation are principal issues that should be considered. Finally, association of the users with BSs becomes an involved problem due to mobility of DBSs. In this paper, we consider a macro-BS (MBS) and several DBSs that rely on the wireless links to the MBS for backhauling. Considering regular and uRLLC users, we propose an algorithm to find efficient 3D locations of DBSs in addition to the user-BS associations and wireless backhaul bandwidth allocations to maximize the sum logarithmic rate of the users. To this end, a decomposition method is employed to first find the user-BS association and bandwidth allocations. Then DBS locations are updated using a heuristic particle swarm optimization algorithm. Simulation results show the effectiveness of the proposed method and provide useful insights on the effects of traffic distributions and antenna beamwidth.