Linear Dispersion Based Research On Space-Time Coding Theory And Applications

MIMO (Multiple Input Multiple Output) technique has drawn great attention because it can greatly enhance the system spectral efficiency or the system performance without increasing the time/frequency resource. Among the academic discussion, STBC (Space Time Block Coding) is always a hot research point for the low computing complexity and ability of achieving diversity gain. Although there are many different methods of studying STBC, the codes can always be analyzed by unified linear dispersion theory. By linear dispersion theory, the design of STBC can be converted to design dispersion matrices, which can combine the algebraic process with the STBC design, in order to make the design process more flexible. This dissertation focuses on the design of STBC based on linear dispersion theory to agree with different transmission scenarios, and the design integratedly considering the transmission performance, transmission rate and the decoding complexity. Moreover, concrete transmission schemes have been proposed based on the proposed STBC designs. The major work and contributions of this dissertation consist in:1. In the first part, the dissertation discusses the linear dispersion STBC design for the requirement of different transmission rate. Firstly, the structure method of Clifford basic matrix set has been discussed, and then a double-element basic matrix set has been proposed in this part. The two types of basic matrix set include matrices with different dimension, based on which OSTBC (Orthogonal STBC) and QO-STBC (Quasi-Orthogonal STBC) with different transmit antennas have been designed. By integrately measuring the objectives, optimal STBC has been chosen to meet the requirement of different rate and different transmit antennas. Not only does the discussion in this part have certain novelty, but it is also the foundation of the whole dissertation. 2. To meet the requirement of open-loop transmission scenario, the dissertation proposes a type of HDFR-QOSTBC (High Dimension Full Rate QO-STBC) based on linear dispersion theory. This type of codes has the characteristics:high dimension, full rate transmission and controllable decoding complexity. The details of the equivalent channel have been investigated and a useful property is revealed that the eigenvectors of the equivalent channel of these codes are constant, and the eigen-values can form a diagonal matrix with certain structure. By using this conclusion, a preprocessing scheme without feedback is proposed, which can greatly reduce the decoding complexity.3. Traditional multimedia broadcasting system is basically relying on equal error protected transmission, which results in that the cell-centre user's rate is limited by the channel environment of the cell-edge user. However, unequal error protected transmission can greatly enhance the entire transmission ability of the broadcasting system. In this part of the dissertation, two types of STBC with unequal error protection (UEP) are proposed. The first type of unequal error protection STBC is designed based on OSTBC or QO-STBC, which superimpose symbols with different performance in the same codeword, and the coding gain can be changed by controlling the power allocation factor. The second type of unequal error protected STBC is termed as UEP-STC, which is designed by linear dispersion theory. In order to achieve unequal error protection, the dispersion matrices corresponding to the transmission symbols of the UEP-STC are grouped, and the number of different groups is not the same. The symbols of UEP-STC belonging to different protected layers can be decoded independently without interference. Moreover, based on UEP-STC, a multilayer multilevel unite transmission scheme which is applicable to the multimedia broadcasting systems is proposed.4. When the code transmission rate reaches the level as multiplexing, the complexity of maximum likelihood decoding will exponentially increase, and STBC design with linear detection is meaningful in this situation. In this part of the dissertation, a type of STBC with multiplexing rate is designed based on linear dispersion theory. The dispersion matrices of the codes are chosen from Clifford matrix set, and it can be proved that this type of codes can reach the optimal performance on the premise of linear detection. Based on the linear optimal STBC, a single-stream adaptive modulation and coding scheme is proposed. This scheme can reduce the time of link adaptive choice, which can greatly reduce the system computing complexity.