消除负频率影响的频谱校正

采用比值法校正低频成分的参数精度比较低，其原因是模型中忽略了负频率成分的贡献。为解决这一问题，基于包含负频率贡献的频谱公式，建立新的校正方案，给出矩形窗和汉宁窗对应的校正公式，采用数值仿真对校正公式进行考核。结果表明，提出的校正公式正确，尤其是汉宁窗的频率定位精度、幅值相对精度和相位精度几乎到双精度运算的精度下限。

SPECTRUM RECTIFYING WITH NEGATIVE FREQUENCY CONTRIBUTION ELIMINATING

The precisions of ratio rectifying method is very rough when it was applied to the case of low frequency component, of which the reason was that the contribution from negative frequency not taken into the model. To overcome this dilemma, the spectrum function of the single frequency signal, which contains the negative frequency contribution, was deduced out. Base on this function, the new rectifying algorithm was proposed, and the rectifying equations were presented for the cases of rectangular window and Hanning window. The new rectifying method was examined by simulation. The study shows that, the presented equations are correct, especially the precisions algorithm corresponding to the Hanning window case almost approaches to the low limit of double precision operations.