Abstract:

Proposed in this paper is an oblique-projection subspace approach to the blind channel estimation of MIMO-OFDM system,which,under conditions of satisfying the identification system,divides the signal-receiving space into three subspaces including the ＂past subspace＂,the ＂present subspace＂ and the ＂future subspace＂.By obliquely projecting the ＂present subspace＂ respectively onto the ＂past subspace＂ and the ＂future subspace＂,the impact of multi-path effect on the receiving system can be effectively eliminated.In addition,a method to estimate the ambiguous matrix of channel is proposed,and the singular value decomposition is adopted to improve the anti-noise ability of the algorithm.It is found that,as compared with the noise subspace approach,the proposed approach is of lower computational complexity only with slight sacrifice of estimation accuracy.Simulated results show the proposed projection approach is efficient for identifying system channel.

Key words: oblique-projection subspace approach, multiple-input multiple-output system, Orthogonal Frequency Division Multiplexing, blind channel estimation, ambiguous matrix