Low-complexity detection and cooperative communication for spatial modulation systems

Spatial modulation (SM) is an alternative method to classical multiple-input multiple-output (MIMO) spatial multiplexing techniques with vertical-Bell Labs layered space-time (V-BLAST) decoding. In SM, the information is transmitted via the conventional amplitude-phase modulation (APM) symbols along with the active antenna indices. Space shift keying (SSK) is a special case of SM where the information is transmitted through only transmit antenna indices. SSK systems are relatively simple; however, their data rate is lower compared to SM for the same number of transmit antennas. Since the SM symbols are conveyed by only one transmit antenna, one radio frequency chain is used at the transmitter and inter-channel interference (ICI) is removed. On the other hand, transmission rate is limited due to the use of single antenna. Generalized spatial modulation (GSM) mitigates this problem by activating more than one transmit antenna. In GSM, information is both transmitted through the combination of transmit antennas and APM symbols. Hence the transmission rate is increased with the use of multiple active transmit antennas. SM techniques have attracted considerable attention from researchers in the past few years and have been considered as potential candidates for next generation wireless networks. Since SM scheme exploits the spatial domain for data transmission, its capacity calculation is different than the classical systems. In SM, the information is conveyed not only through the M-ary constellation domain but also through the antenna domain, the capacity of SM is expressed as the sum of the capacities of these two domains. A detailed analysis of the outage probability of SM has not been given in the literature yet. In this thesis, first, we derive the outage probability performance of the classical SM system. It is shown that SM systems provide better performance compared to conventional modulation systems in terms of outage probability. The optimum performance of SM can be achieved with the maximum likelihood (ML) signal detection but computational complexity increases with exhaustive search. Since the inter-antenna interference arises during the transmission, GSM receiver is much more complicated than the SM scheme. In this work, we investigated a novel low complexity detection algorithm for SM systems (especially for GSM systems due to their receiver complexity). The algorithm can be split into two stages: in the first stage, least squares (LS) estimate of the transmitted signal is used in the ML detection to find the sub-optimal solution of the antenna combinations. The N best estimate of antenna combinations are selected in order to reduce the search space. In the second stage, this reduced set and the LS estimates of transmitted symbols based on this set are then sent to the optimal ML detector to find the correct antenna combinations and APMsymbols. It is shown that the proposed algorithm obtains near ML performance with a lower computational complexity. Cooperative communication has attracted numerous researchers over the past decade. In a cooperative scenario, a source (S) transmits its data to the relay (R) and the destination (D) in the first phase and the relay forwards the source's information either decoding the received signal (decode-and-forward, DF) or amplifying it (amplify-and-forward, AF) in the second phase. This forwarding concept forms a virtual MIMO system to combat fading and it is very effective to gain a larger coverage. In this thesis, the advantages of SM and cooperative communication systems have been combined and the outage probability analysis of SM system is extended to the cooperative scenarios under some relaying techniques. Cooperative SM systems are also compared to conventional modulation systems and provide better performance in terms of outage probability. Finally, we propose novel cooperative SM systems with AF and DF techniques where all nodes have multiple transmit and/or receive antennas, an issue which has not been studied before. The previous studies in the literature for cooperative SM systems generally consider the SSK technique instead of SM. Additionally, most of these studies assume single transmit and/or receive antenna at the relay(s) and destination. As known, the SM/SSK schemes need at least two transmit antennas to map information bits to the antenna indices. Furthermore, a cooperative SM system with DF relaying where the relay(s) has only one transmit antenna is not a complete SM system since the relay(s) can not re-encode the decoded data into the SM symbols. Moreover, to improve the error performance compared to APM, an SM system requires at least two receive antennas. To the best of our knowledge, a comprehensive work on cooperative SM systems that have multiple transmit and/or receive antennas at R and D has not been performed in the past yet. In this work, we derive the average bit error probabilities (ABEP) for both the MIMO-AF and the MIMO-DF systems and validate them with the computer simulations results. Furthermore, these two cooperative SM systems are compared with the classical cooperative system using M-ary modulations in terms of BER performance. Computer simulations and analytical expressions show that the proposed MIMO-DF and MIMO-AF systems provide considerable error performance improvements over conventional cooperative APM systems. Lastly, the BER comparison of MIMO-AF and MIMO-DF cooperative SM systems are presented.