Quasi Orthogonal Space-time Block Codes Structure And Performance Analysis

In recent years, with the popularity of mobile communications and the rapid development of related businesses, worldwide spectrum resources becomes increasingly tense. The space-time coding techniques integrare effectively the advantage between time diversity and spatial diversity, which contributes to high communication data transmission efficiency based on the diversity gain, and relieves effectively the stress of the spectrum resources. Quasi-orthogonal space-time block codes have the advantage in full rate transmission although in the expense of orthogonality. This dissertation focuses on the encoding and decoding of quasi-orthogonal space time block codes. A new construction of quasi-orthogonal space-time block code and decoding algorithm of the improved SD were proposed, based on the deep research and analysis of QOSTBC.Firstly, based on coding theory of space-time coding and the characteristics of quasi-orthogonal space time block codes, we proposed a four transmit antenna QOSTBC. Then we obtain an improved QOSTBC using the Givens matrix rotation method to preprocessing the transmitted symbols, which has the better performance than traditional one. Simulation results show that the bit error rate of the new QOSTBC has been improved significantly.Secondly, based on the TBH encodes, a constellation rotation TBH encodes is proposed, which eliminates interference brought by the non-orthogonality by matrix rotation. Compared with the traditional QOSTBC, constellation rotation TBH reduces the decoding complexity. The simulation results show a significant improvement in the BER performance than the constellation rotation TBH code.Finally, according to the principle of the VB sphere decoder, an improved VB-SD is presented in this dissertation, which is with lower complexity than the VB-SD. For a large constellation, the simulation results shows the obvious improvement in the search timeof the new algorithm. Especially in the low SNR case, the proposed VB-SD not only improve the system response speed, but also reduces the decoding complexity.