Abstract: The electrocardiography signal includes noise and has high dimension due to the acquisition of equipment, testing the mood, in order to improve the identification accuracy, this paper proposes a new ECG signal identification method based on wavelet transform and support vector machine.Firstly, wavelet transform is used to eliminate the nose of ECG signal, and then the correlation analysis method is used to reduce the feature dimension;lastly, support vector machine is used to build ECG signal identification model.The simulation results show that the method improves the identification accuracy and speed, it is efficient identification algorithm.