Low-Complexity ML Detection for Spatial Modulation MIMO With APSK Constellation

Abstract

Spatial modulation (SM) is a promising multiple-input–multiple-output (MIMO) transmission technology, which exploits the indexes of the transmit antennas for encoding information into the spatial dimension and, thus, for increasing the transmission rate. Optimal demodulation is realized by using a maximum-likelihood (ML) detector, which jointly estimates the information bits encoded into the transmit antenna indexes and into the transmitted symbol. If a large number of transmit antennas and/or a large modulation size are used; however, the computational complexity of ML demodulation may not be affordable for some applications. Against this background, we propose two novel near-ML but low-complexity detection schemes for SM-MIMO systems, which use $\boldsymbol{M}$ -ary amplitude and phase-shift keying (APSK) digital modulation. The proposed algorithms exploit specific features of the $\boldsymbol{M}$ -ary APSK constellation diagram, in order to avoid searching among all constellation points. It is shown, in particular, that the complexity of the proposed detection schemes is independent of the constellation size and that it only depends on the number of rings that constitute the APSK constellation. Simulation results are also provided, which confirm that the proposed detection schemes provide bit error probability performance close to that of ML demodulation but with lower demodulation complexity.