在采用独立时频块导频方案的MIMO-OFDM系统信道估计中,证明了导频数小于多项式模型参数时,使用标准的广义逆进行多项式内插无法得到信道估计的最佳值。采用基于最小方差准则构造的加权矩阵对广义逆的定义做出修正,提出了基于加权广义逆矩阵的二维多项式内插算法。仿真结果证明该算法可以适应不同的移动速度,特别是在中低移动速度下,该算法在高信噪比时的误码性能相对于一维线性内插算法可以得到约2 dB的增益。
For adopting single time-frequency chuck as pilot pattern.We prove that the optimal estimation cannot be achieved by polynomial interpolation using the standard inverse matrix,when there are less pilot symbols than the parameters of the polynomial model.For solving this proldem,a generalized inverse matrix is defined from a weighting matrix which is constructed based on minimum variance criterion.A 2-D polynomial interpolation algorithm is then proposed based on this weighted generalized inverse matrix.Simulation result shows that the proposed algorithm can be applied to receivers with different velocities.Under moderate or low velocity in particular,this algorithm can achieve a BER gain of about 2 dB for large SNR compared with 1-D linear Interpolation algorithms.