ICIC-Enabled Association and Channel Selection for UAV-BSs Based on User Locations and Demands

In this work, the downlink association problem is studied when inter-cell interference coordination (ICIC) is enabled at the network side. The serving base stations (BSs) are mounted over unmanned aerial vehicles (UAVs), allowing for their initial placement as per the user geographic distribution. With recent air-to-ground channel modeling adopted herein, we focus on addressing the problem of coordinated channel assignment from within a reusable channel pool at each of the operator-deployed UAV-BSs along with the consideration of backhaul constraint. The joint association/channel assignment problem, which enables a flexible design of the inter-cell interference terms experienced by the users, is cast into a mixed integer linear program (MILP) that is numerically solved. Due to being a non-deterministic polynomial time (NP)-hard problem, the computation time may unboundedly grow, especially as the number of network nodes increases. To solve this problem, a low-complexity heuristic algorithm is proposed. The proposed heuristic solution is compared to that obtained via the MOSEK solver, revealing a close achievable performance with much less computational complexity.