The FXLMS algorithm is widely used in active vibration control system. The estimation of secondary path plays very important roles in such a system. This paper presents an experimental investigation of effective secondary path estimation in active vibration control using measured Frequency Response Function (FRF). Principal component analysis (PCA) is pursued to the measured FRF for noise elimination, and then the PCA-compressed FRF data are used for secondary path estimation. The control results indicate that the proposed method has good control performance.