The Electrocardiogram (ECG) is essential for the clinical diagnosis of cardiovascular disease. An advanced algorithm combining wavelet packet transformation (WPT) and Hilbert Huang transform (HHT) is presented for processing ECG (Electrocardiography) signal in this paper. First the WPT can resolve the ECG signal into a group of signals with narrow band. Then, the Empirical Mode Decomposition (EMD) process of Hilbert-Huang Transform (HHT) is applied on the narrow band signals. The unrelated IMFs of ECG signal are removed from result through a screening process. Finally, the Hilbert transform is employed to achieve the Hilbert spectrum and marginal spectrum. The results show the effective performance of the algorithm combining WPT and HHT in reducing ECG noise and time-frequency analysis. By comparing with the original HHT, the proposed algorithm has the better performance on ECG signal processing. © Springer Science+Business Media Dordrecht 2014.