Radio resource allocation for heterogeneous traffic in GFDM-NOMA heterogeneous cellular networks

摘要

In this study, the authors consider the downlink radio resource allocation for heterogeneous traffic in generalised frequency division multiplexing (GFDM)-non-orthogonal multiple access (NOMA) based heterogeneous cellular networks. In this scheme, multiple number of users can be allocated on each subcarrier. Two types of traffic are considered, elastic and streaming. The problem of maximising the weighted sum-rate of elastic users is addressed subject to streaming users minimum rate in addition to subcarrier and transmit power constraints. This problem is a non-convex NP-hard optimisation problem. To solve this problem, the authors divide it into two subproblems, subcarrier allocation and power allocation then an iterative algorithm is proposed. Subcarrier allocation is updated by solving an integer linear program, where a successive convex approximation approach is adopted to transform the power allocation subproblem to a sequence of convex subproblems, using one of the three methods, successive convex approximation for low ComplExity, arithmetic-geometric mean approximation (AGMA) and difference of two concave functions to find the power allocation optimal solutions. Numerical experiments show that the proposed algorithms can improve the system performance. Furthermore, they show that AGMA can achieve a sum-rate near to the global optimal solution, at the expense of more computational time.