Research Of Blind Equalization Algorithm In MIMO-OFDM System

Modulation technique which can make full use of bandwith and provide high data transfer rate is always a focus in wireless communication. It is considered to increase the system capacity and inhibit the channel fading effectively with the application of multiple-input and multiple-output(MIMO) technique. When the transmission bandwith and power is constant, it improves the spectral efficiency and reduces the error rate. MIMO technique is always used in conjunction with orthogonal frequency division multiplexing, which evolves from the early concept of frequency division multiplexing and is applied in broadband wireless communication field. This modulation technique can improve the transmission rate, eliminate inter symbol interferance and meanwhile improve the spectral efficiency.Traditional channel identification method for MIMO-OFDM systems has ineffective transmission due to the large amount of redundant training sequence for time-varying channel, and it is inevitable to eliminate the ambiguity by the blind method based on subspace.In this paper, MIMO, OFDM and other related technologies are introduced firstly and a common MIMO-OFDM system module is established. After listing several existing channel identification methods under similar module and simulation, demands for blind identification in MIMO-OFDM system are discussed. In subspace method, autocorrelation matrix of received vectors is calculated first, then the estimation of channel parameters is obtained using the orthogonality between signal subspace and noise subspace. Then an estimation method exploiting the ambiguity matrix is proposed on basis of subspace method, which using much less training symbols and maintaining the precision. The simulation results show the effectiveness of the algorithm.