Improved Spatial Modulation Techniques for Wireless Communications

Transmission and reception methods with multiple antennas have been demonstrated to be very useful in providing high data rates and improving reliability in wireless communications. In particular, spatial modulation (SM) has recently emerged as an attractive transmission method for multiple-antennas systems due to its better energy efficiency and lower system complexity. This thesis is concerned with developing transmission techniques to improve the spectral efficiency of SM where antenna/subcarrier index involves in conveying information bits. In the first part of the thesis, new transmission techniques are developed for SM over frequency-flat fading channels. The first proposed scheme is based on a high-rate space-time block code instead of using the classical Alamouti STBC, which helps to increase the spectral efficiency and achieve a transmit diversity order of two. A simplified maximum likelihood detection is also developed for this proposed scheme. Analysis of coding gains and simulation results demonstrate that the proposed scheme outperforms previously-proposed SM schemes at high data transmission rates. Then, a new space-shift keying (SSK) modulation scheme is proposed which requires a smaller number of transmit antennas than that required in the bi-space shift keying (BiSSK). Such a proposed SSK-based scheme is obtained by multiplexing two in-phase and quadrature generalized SSK streams and optimizing the carrier signals transmitted by the activated antennas. Performance of the proposed scheme is compared with other SSK-based schemes via minimum Euclidean distance analysis and computer simulation. The third scheme proposed in this part is an improved version of quadrature SM (QSM). The main feature of this proposed scheme is to send a second constellation symbol over the in-phase and quadrature antenna dimensions. A significant performance advantage of the proposed scheme is realized at the cost of a slight increase in the number of radio-frequency (RF) chains. Performance comparisons with the most recent SM schemes confirm the advantage of the proposed scheme. The last contribution of the first part is an optimal constellation design for QSM to minimize the average probability of error. It is shown that, the error performance of QSM not only depends on the Euclidean distances between the amplitude phase modulation (APM) symbols and the energies of APM symbols, but also on the in-phase and quadrature components of the QSM symbols. The analysis of the union bound of the average error probability reveals that at a very large number of transmit antennas, the optimal constellations for QSM converge to a quadrature phase shift keying (QPSK) constellation. Simulation results demonstrate the performance superiority of the obtained constellations over other modulation schemes. In the second part of the thesis, the applications of SM in frequency-selective fading channels are studied. First, a new transmission scheme that employs SM for each group of subcarriers in orthogonal frequency-division multiplexing (OFDM) transmission is investigated. Specifically, OFDM symbols in each group are passed through a precoder to maximize the diversity and coding gains, while SM is applied in each group to convey more information bits by antenna indices. Performance analysis and simulation results are carried out to demonstrate the superiority of the proposed scheme over a previously-proposed combination of SM and OFDM. Next, the performance of OFDM based on index modulation and a flexible version of OFDM, knows as OFDM with multiple constellations, is compared for both case of "no precoding'' and "with precoding'' of data symbols. It is shown that the precoded OFDM with multiple constellations outperforms precoded-IM based OFDM systems over frequency-selective fading channels. The last part of the thesis investigates a multiuser downlink transmission system based on in-phase and quadrature space-shift keying modulation and precoding to reduce the minimum number of transmit antennas while keeping the complexity of the receiver low. In addition to the maximum likelihood (ML) detection, the low complexity zero forcing (ZF) receiver is also studied. Theoretical upper bounds for the error probabilities of both ML and ZF receivers are obtained and corroborated with simulation results