Sparse Multi-User Detection For Non-Orthogonal Multiple Access In 5g Systems

A key technology in the 5th generation (5G) wireless communications is non-orthogonal multiple access (NOMA) which can effectively support massive random access. Since in a 5G system the number of active users generally does not exceed 10% of the total number of users in practice, the main challenge for multi-user detection (MUD) is the joint detection of sparse active users and their data. To address this challenge, we propose a novel compressed sensing based NOMA (CS-NOMA) scheme which can be deployed in a CDMA or an OFDM system and does not have control signaling overhead and need no knowledge of the active state of users. In the proposed scheme, K users are multiplexed over N shared resources (K >> N). Then, a CS-based MUD is applied to the proposed CS-NOMA scheme. Simulation results show that, in the case that active users are sparse (e.g., 10%), the proposed scheme can achieve a superior performance even when the system loading is up to 300%.