Adaptive Space Modulation With Partial CSIT in Spatially Correlated Fading Channels

The problem of adaptive space modulation (ASM) design using partial channel state information at the transmitter (CSIT) is investigated. ASM relies on adaptive modulation of information symbols in the spatial domain using multiple antennas based on some sort of CSIT. The CSIT is exploited to adjust the transmission such that the distances between the receive constellation vectors are increased to reduce the probability of error. We assume that the spatial correlation matrices are available as the partial CSIT and introduce two different approaches to design ASM. In the first approach, we consider that all the transmit antennas are active at each transmit interval and, using an upper bound on the average probability of error, design the transmit vectors based on the spatial correlation matrices to improve the performance. In the second approach and to avoid interantenna synchronization, we assume that only one transmit antenna is active at each transmit interval, and we modify a space shift keying (SSK) transmission by transmitting different complex values from different transmit antennas. Finally, we derive a power allocation to improve further the performance in nonstationary channels. By simulation, we show the effectiveness of the proposed approaches in improving the performance in terms of the symbol error rate (SER).