Low-Complexity MIMO-SCMA Detection for LEO Satellite Communications

In order to unlock the full advantages of low earth orbit (LEO) satellite communications, non-orthogonal multiple access (NOMA) has been introduced for supporting massive access in the uplink. However, the high detection complexity prevents NOMA from practical implementation. Accordingly, in this work, we consider the accuracy-complexity trade-off for the uplink multiple-input multiple-output (MIMO)-NOMA in satellite communications and propose an improved approximate message passing (IAMP) detection algorithm. We further modify our IAMP to adapt to imperfect ground-space channels. Simulation results demonstrate that the proposed IAMP detector achieves a lower bit error rate (BER) performance than the low-complexity AMP detector, approaching the state-of-the-art message passing algorithm (MPA) detector, and attains a lower detection complexity than both MPA and AMP detection.