Study On Hybrid Space-Time Coding Based On Uniform Channel Decomposition Precoding

Due to the low minimum diversity and decoder's error propagation, the vertical Bell-Labs layered space-time (VBLAST) MIMO system has the degradation in the bit error rate (BER) performance. Thus, a hybrid system combined the orthogonal space time block codes (STBC) with a VBLAST system. However, if the VBLAST is precoded using the channel state information at the transmitter, the hybrid system's performance may be better. This motivates this thesis apply the uniform channel decomposition (UCD) precoing technology to the STBC-VBLAST hybrid system, and the simulation results indicate that the precoded hybird space-time coding system has a better perfomance than the unprecoded hybrid sustem.The research background of this issue and the home and abroad research status are reviewed in this thesis. Then the main features of the MIMO technology and the system model are discussed, the channel capacity of the MIMO systems is described in detail, and the analysis of the water-filling based on singular value decomposition (SVD) lays the foundation of the research of precoding techology. It is known through the simulation and the analysis that the precoding techology can increase system channel capacity efficiently, espesally in the case of low signal noise ratio.After that, this thesis highlights the space-time block coding and layered space-time coding, carefully studys the encoding and decoding of the Alamouti coding scheme, and analys the SNR of the Alamouti system under the condition of different diversities. Through a large number of comparative simulations of the performance of various decoding methods, we analyze the LSTC decoding.This precoding algorithm is applied to STBC-VBLAST hybrid space-time coding systems in this thesis, which is lossless of capacity, and also can improve the diversity gain, to achieve the best compromise between diversity and multiplexing, and introduces a new STBC-VBLAST coding scheme based on the UCD precoding algorithm. Theoretical analysis and simulation results indicate that the scheme proposed in this thesis can reach an ideal BER performance.Finally, this thesis is summarized and the future is forecasted.