On Power Allocation and User Grouping for Sparse Coded Non-Orthogonal Multiple Access in the Downlink

Sparse-spreading based non-orthogonal multiple access is considered to be a key component to support massive connectivity and increased capacity for future wireless communication systems. Since multiple users are co-scheduled on the same physical resources, power allocation cannot be optimized independently per user but also has to consider the assignment of resources between different users. This accounts specifically in the downlink, where the sum power constraint holds. In this paper, we propose a heuristic approach for power allocation in a downlink system using multi-user sparse coded multiple access. The motivation is to improve fairness within a group of users sharing the same physical resources in a non-orthogonal fashion and thus, simplify the receiver processing. An extensive simulation study reveals, that aligning the signal-to-noise ratio of each user within the group, user fairness can be improved without loosing system performance as long as the users are grouped carefully.