Constellation design and performance analysis of the parallel spatial modulation

Abstract

SummarySpatial modulation (SM) is a relatively recent multiple‐input multiple‐output (MIMO) system in which information is carried by the index of the antenna used for transmission as well as by the conventional signal symbols. Several systems that build upon SM have since been proposed including the generalized SM (GSM), a variant of GSM with multiple active antennas (MA‐SM), quadrature SM (QSM), and parallel SM (PSM), among others. The PSM system can increase the spectral efficiency by splitting the antenna set into groups and applying SM independently in each group using the same signal symbol. In this paper, we first derive the upper bound on the error probability of the PSM. The search of the optimal constellation set is then formulated as a multi‐objective optimization problem, where the obtained constellation minimizes the asymptotic error probability. We conclude that as the number of antenna groups increases, the proposed constellation converges to the conventional phase‐shift keying at relatively low number of transmit antennas. The simulation results show that the proposed constellation outperforms conventional constellations by as much as 5 dB, for high‐modulation orders. Since the multi‐objective optimization is independent of the channel matrix, it can be easily done off‐line. This implies that these gains come at no complexity or delay cost.