A sparsity adaptive multi-user detection algorithm for SIMO-NOMA systerms

Abstract

Non-orthogonal multiple access(NOMA) can improve spectrum efficiency and support massive connectivity by the use of resources in non-orthogonal way, which is expected to become one of the key technologies of 5G. Considering the situation that the base station(BS) is equipped with multiple antennas，this paper proposes a compressive sensing(CS) based sparsity adaptive matching pursuit hard fusion algorithm(SAMP-HFA) to realize multi-user detection(MUD) for uplink grant-free single-input multiple-output non-orthogonal multiple access(SIMO-NOMA) systems where the number of active user is unknown. The proposed algorithm consists of three steps. Firstly, it detects the user activity information by conventional SAMP algorithm at each antenna, and then amalgamates the detected user activity information to obtain a common active user set. Finally, the users′ data can be detected by the obtained active user set. The results show that the proposed SAMP-HFA demonstrates significant performance gain in terms of bit error rate(BER) with the number of antennas increases.

Keywords

• compressive sensing
• multi-user detection
• SIMO-NOMA
• sparsity adaptive
• hard fusion