Joint Link Selection and Resource Allocation for Aeronautical Heterogeneous Networks

Comprising both air-to-ground data links and satellite communications, the aeronautical telecommunication network (ATN) is inherently heterogeneous. However, the potential of the ATN has not been fully exploited, predominantly due to the suboptimal resource provisioning strategy, which overlooks the link-switching overhead and unbalanced cost trade-offs. This paper aims at providing a cost-minimization strategy for airborne terminals, by jointly designing the link selection and the resource allocation. Specifically, in contrast to the state-of-the-art on the ATN, where link-switching costs are ignored, a network economics trade-off between the expense paid to communication service providers and the link switching overhead is formulated as a combinatorial optimization problem, whilst guaranteeing the performance of the airborne queue backlog. With the aid of the Lyapunov optimization framework, the original stochastic problem is decomposed into a set of sub-problems at each temporal interval, while ensuring punctual data transmission by introducing the virtual queues. The problem is posed as an NP-hard mixed integer nonlinear programming, and hence it is difficult to find its solutions within polynomial time using the classic optimization approaches. Alternatively, a heuristic solution is then provided as the link selection and resource allocation strategy. Our extensive numerical results demonstrate the superiority of the strategy proposed. It has been confirmed that the proposed strategy achieves near-optimal cost performance, simultaneously balancing service provider expenses and network switching overhead at a low computational complexity of <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex-math notation="LaTeX">$\mathcal {O}(T \cdot K)$</tex-math></inline-formula>, especially under the <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex-math notation="LaTeX">$N \ll K$</tex-math></inline-formula> condition, while maintaining costs merely 6.3% higher than the commercial optimizer Gurobi.