Distributed Space-Time Block Codes in Wireless Cooperative Networks
cooperative systems , space-time codes , diversity methods , fading channels
In cooperative networks, relays cooperate and form a distributed multi-antenna system to provide spatial diversity. In order to achieve high bandwidth efficiency, distributed space-time block codes (DSTBCs) are proposed and have been studied extensively. Among all DSTBCs, this thesis focuses on the codes which are single-symbol maximum likelihood (ML) decodable and can achieve the full diversity order. This thesis presents four works on single-symbol ML decodable DSTBCs. The first work proposes the row-monomial distributed orthogonal space-time block codes (DOSTBCs). We find an upper bound of the data-rate of the row-monomial DOSTBC and construct the codes achieving this upper bound. In the second work, we first study the general DOSTBCs and derive an upper bound of the data-rate of the DOSTBC. Secondly, we propose the row-monomial DOSTBCs with channel phase information (DOSTBCs-CPI) and derive an upper bound of the data-rate of those codes. Furthermore, we find the actual row-monomial DOSTBCs-CPI which achieve the upper bound of the data-rate. In the third and fourth works of this thesis, we focus on error performance analysis of single-symbol ML decodable DSTBCs. Specifically, we study the distributed Alamouti's code in dissimilar cooperative networks. In the third work, we assume that the relays are blind relays and we derive two very accurate approximate bit error rate (BER) expressions of the distributed Alamouti's code. In the fourth work, we assume that the relays are CSI-assisted relays. When those CSI-assisted relays adopt the amplifying coefficients that was proposed in  and widely used in many previous publications, upper and lower bounds of the BER of the distributed Alamouti's code are derived. Very surprisingly, the lower bound indicates that the code cannot achieve the full diversity order when the CSI-assisted relays adopt the amplifying coefficients proposed in . Therefore, we propose a new threshold-based amplifying coefficient and it makes the code achieve the full diversity order two. Moreover, three optimum and one suboptimum schemes are developed to calculate the threshold used in this new amplifying coefficient.