Mimo-ofdm System Channel Blind Estimation Algorithm

Multiple Input Multiple Output and Orthogonal Frequency Division Multiplexing (MIMO- OFDM) has been considered as the key technology for the next generation wireless communication systems due to its high rate transmission capability, hign spectrum efficiency and its robustness to frequency-selective fading. Channel estimation is an important part for coherent detection. Channel can often be estimated by training sequences or aid of pilots, but they waste rate bandwidth and reduce the efficiency of systems. Because of saving bandwidth and being capable of tracking slow channel variations, blind channel estimation and equalization methods are well motivated as they avoid of training sequence and pilot.Firstly in this thesis, the properties of the wireless fading channel are introduced. Secondly, the basic principle and realization of MIMOOFDM systems are introduced, and the system construction and system model are described. An improved algorithm is proposed based on the conventional channel blind estimation algorithm which uses the auto- correlation matrix of the vector of received signal in the OFDM systems. The improved algorithm uses cross-correlation between cyclic prefix and a special sub-vector from vector of received signal to implement channel blind estimation. The improved algorithm not only reduces the complexity of the conventional algorithm, but also improves its performance in theory.