Channel Estimation Method Based On Two-Dimensional Pilot In MIMO-OFDM System

This thesis mainly studies the channel estimation technique in MIMO-OFDM system, which is the crucial technology of the next generation mobile communication. Channel estimation technique is a kind of estimator. It estimates the channel impulse response with channel state information parameters by multi-path instantaneous fading in the radio channel. Precisely estimation ensures the perfect performance of the transmission progress in MIMO-OFDM system. Hence, channel estimation is one of the most important technologies for wireless communication as a hotpot.First, in the thesis we investigate the multi-path fading characteristics with the wireless fading parameters of the mobile communication channel. After analyzing the characteristics, study the basic principles of MIMO-OFDM system. Furthermore, the channel estimation technique in MIMO-OFDM system is investigated, with the emphasis on pilot-based channel estimation. Both the traditional and improved algorithms are given respectively, and the analysis and comparisons of these algorithms are made by Matlab simulation in the end.Second, this thesis put emphasis on the promotional performance of the estimator based on two-dimensional pilot. According to the time-varying fading characteristics, analyses the channel estimation methods with two-dimensional pilot in details. Furthermore, propose the novel advanced algorithms to enhance the estimator performance by pilot patterns, the estimation algorithms and the noise-reducing filter under DFT/IDFT transform. After the comparison of the traditional and improved algorithms by computer simulations, validate the algorithm's effectiveness and show that the method is feasible. The problem of large computational complexity caused by inversion of auto-covariance matrix is also solved.At the end of this thesis, we study a novel estimation algorithm based on superimposed pilot sequences. This method utilizes the pilot's auto-correlation function and the cross-correlation function of pilot and received data sequence to achieve the goal. Then, deeply analyze the advantages of the superimposed sequences, proposed a novel channel estimation method for the entire channel impulse response utilizing iterative algorithm. And analyze the influence of iteration times to the system performance. In the end, presents a comprehensive prospect of this technology's future.