Resource Allocation for Cognitive Satellite-HAP-Terrestrial Networks With Non-Orthogonal Multiple Access

Cognitive radio and non-orthogonal multiple access are promising to improve the spectrum efficiency for the satellite-high altitude platform (HAP)-terrestrial networks. To achieve the optimal system performance and guarantee the quality of service, we formulate a constrained mixed integer nonlinear programming problem to maximize the sum rate of secondary network for the considered NOMA-enabled cognitive satellite-HAP-terrestrial network, while considering the constraints of the quality of service of primary users, the maximum power of HAP, and the number of secondary users in a NOMA group. To tackle the non-convex optimization problem, we decouple it into subchannel assignment and power assignment subproblems. Then, a greedy heuristic algorithm is proposed to assign subchannels to secondary users. Moreover, to solve the power allocation subproblem, we further design a power allocation algorithm by utilizing successive convex approximation, dual decomposition, and subgradient methods. On this basis, an iterative joint resource allocation algorithm is designed. Numerical simulations and comparisons with the orthogonal multiple access scheme are provided to verify the effectiveness of the proposed scheme.