为提高OFDM系统的传输效率,提出了一种利用接收信号的二阶统计特性实现信道盲估计的子空间算法.首先将OFDM信号分成3部分,利用循环前缀(CP)引入的信息冗余,对OFDM系统的传输方程作矩阵变换,得到一个信道矩阵为Toeplitz矩阵的新方程.基于此方程推导出信道估计的子空间算法.Toeplitz矩阵结构使算法的推导和实际的计算大为简化.该算法不需要改变OFDM系统结构,不受信道零点位置的限制,在信道过估计的情况下也适用.实验结果验证了算法的有效性. In order to improve the transmission efficiency of OFDM system, a sub-space algorithm for channel blind estimation based on the second-order statistical characteristics of received signal is proposed.Firstly, the OFDM signal is divided into three parts and the information redundancy introduced by cyclic prefix (CP) The transmission equation of OFDM system is transformed by matrix, a new equation of channel matrix is ​​obtained, and a subspace algorithm of channel estimation is deduced based on this equation.Tootlitz matrix structure simplifies the derivation of the algorithm and the actual calculation, It is necessary to change the structure of the OFDM system without being limited by the zero position of the channel and is also applicable in case of channel over-estimation. Experimental results verify the effectiveness of the algorithm.