In view of the demand of monitoring the impact and vibration of laminated composite plates, a new method of low speed impact identification based on wavelet packet decomposition and distributed fiber Bragg grating sensing network is proposed. The sensor network layout is designed according to the load form of clamped plate structure and the sensing characteristics of fiber Bragg grating sensor. Then, the impact response signals monitored by fiber Bragg grating sensor network are analyzed by the fast Fourier transformation and wavelet packet decomposition to obtain the time domain characteristics of impact characteristics. On this basis, the mutual correlation coefficient between each characteristic decomposition signal and its corresponding time domain original signal is calculated to be the weight of the similarity. The characteristics of the impact response signals of all samples are decomposed, and the sample database is constructed. Finally, the location coordinates of impact points are determined according to the similarity calculated based on the Haudorff distance between the test signal and the sample information. The research shows that this method can realize the identification of the low speed impact position on the structure of the laminated plates.