In the process of metal milling, especially in the machining of low?stiffness workpieces, chatter is a key factor affecting many aspects such as surface quality, machining efficiency and tool life. In order to avoid the chatter, a milling chatter recognition method based on dynamic and Wavelet Packet Decomposition(WPD) is proposed from the signal processing. The modal parameters of the system are obtained by modal experiments. Based on the principle that the chatter frequency will peak near the natural frequency of the system, the original milling force signal is decomposed by WPD, and the sub?signals containing rich chatter information are selected for signal reconstruction. Finally, the time?frequency and Hilbert spectrum characteristics of the reconstructed signal are compared and analysed, and the chatter recognition is performed. At last, the proposed method is verified by the experiments. The results demonstrate the effectiveness and reliability of the proposed method.