Optimal SIC Ordering and Power Allocation in Downlink Multi-Cell NOMA Systems

Abstract

In this work, we consider the problem of finding globally optimal joint successive interference cancellation (SIC) ordering and power allocation (JSPA) for the general sum-rate maximization problem in downlink multi-cell NOMA systems. We propose a globally optimal solution based on the exploration of base stations (BSs) power consumption and distributed power allocation. The proposed centralized algorithm is still exponential in the number of BSs, however scales well with larger number of users.For any suboptimal decoding order, we address the problem of joint rate and power allocation (JRPA) to achieve maximum users sum-rate. Furthermore, we design semi-centralized and distributed JSPA frameworks with polynomial time complexity. Numerical results show that the optimal decoding order results in significant performance gains in terms of outage probability and users total spectral efficiency compared to the channel-to-noise ratio (CNR)-based decoding order known from single-cell NOMA.Moreover, it is shown that the performance gap between our proposed centralized and semi-centralized frameworks is quite low. Therefore, the low-complexity semi-centralized framework with near-to-optimal performance is a good choice for larger number of BSs and users.

Optimal Water-Filling Algorithm in Downlink Multi-Cluster NOMA Systems

“…The strong duality holds since (4c) and (4e) hold with strict inequalities. In the KKT optimality conditions analysis in Appendix C of [7], we proved that in fully SC-SIC, the maximum power budget does not have any effect on the optimal powers obtained in the power minimization problem. And, at the optimal point, all the multiplexed users get power to only maintain their minimal rate demands.…”

“…Moreover, g n k is the (generally complex) channel gain from the BS to user k on subchannel n, and z n k ∼ CN (0, σ n k ) is the additive white Gaussian noise (AWGN). We assume that perfect channel state information (CSI) of all the users is available at the schedular [7], [8].…”

“…For any given feasible q, (4c) and (4e) can be removed from (4). Then, the convex problem (4) can be equivalently divided into N intra-cluster convex power allocation subproblems, where for each subchannel n, we find the optimal powers according to Appendix B in [7]. It is proved that at the optimal point of sum-rate maximization problem in fully SC-SIC, all the users with lower decoding order get power to only maintain their minimum rate demands.…”

“…In contrast to fully SC-SIC in [7], in NOMA, there is a competition among cluster-head users to get the rest of the cellular power. According to (7), the optimal power of the NOMA cluster-head user Φ n can be obtained as a function of q n given by…”

“…The optimal power allocation in NOMA includes two components: 1) optimal power allocation among multiplexed users within each cluster; 2) optimal power allocation among these isolated clusters (optimal power budget of clusters). In our previous work [7], we showed that the Karush-Kuhn-Tucker (KKT) optimality conditions analysis for fully SC-SIC, also called single-cluster NOMA, results in a unique closed-form of optimal powers for both the sum-rate maximization and power minimization problems. In this work, we consider the general NOMA system with multiple clusters each having any arbitrary number of users.…”