Study On Synchronization And Channel Estimation In MIMO-OFDM System

Multiple-input multiple-output(MIMO) and orthogonal frequency division multiplexing (OFDM) technologies become research hotspot now and are known as the most promising physical technologies for B3G/4G broadband wireless mobile communication systems, because of their high spectrum utilization and the reliable broadband data services they provide.On the basis of studying the most fundamental theory of MIMO-OFDM wireless communication system, this dissertation studies synchronization and channel estimation in MIMO-OFDM system in detail. The major achievements are outlined as follows:1.The channel mathematic models and modeling method of wireless mobile channel are studied, especially MIMO coherence channel model.2.The timing synchronization and frequency synchronization algorithms based on training sequence in MIMO-OFDM system are studied. Meanwhile, on the basis of the work above, this paper designs the synchronization scheme of MIMO-OFDM WLAN, which utilizes orthogonal training sequence. And the delay auto-correlation is exerted in timing synchronization while the correlation of the preamble in time domain and frequency domain is exerted in frequency synchronization. Analysis and the simulation results show the effectiveness and low implementation complexity of the synchronization scheme.3.A novel frame synchronization algorithm is presented in order to improve the effectiveness of ordinary frame synchronization algorithm. Comparing to the nine short traning length plain mode of the unitary decision value of ordinary frame synchronization algorithm, the novel frame synchronization algorithm obtains the triangle mode of the unitary decision value. It is easy to find peak value using the novel frame synchronization.4. The channel estimation based on training sequences and pilot signals in MIMO-OFDM system are studied. With regard to the former one, on the basis of the discussion on the design of training sequences, LS in time and frequency domain,LMMSE and low complexity LMMSE methods based on SVD with training sequences are investigated..The complexity of computing and implement are analyzed and compared. For the channel estimation scheme based on pilot, the orthogonal design of pilot symbol and several interpolation algorithms are introduced at first. And then linear interpolation,Gaussian interpolation and transform-domain processing with LS method in frequency domain are simulated. Finally, the complexity of computing and implement are analyzed and compared.The research in the dissertation indicates the anticipative result. Comparing to ordinary frame synchronization algorithm, the presented frame synchronization algorithm is easy to detect the frame advent. And the synchronization scheme of MIMO-OFDM WLAN is effective. With regard to the channel estimation in MIMO-OFDM system, this paper studies the design of the training sequences and pilot signals, estimation algorithms and interpolation algorithms. All of these works provided reference significance for scheme selection of channel estimation in MIMO-OFDM system.