An ECG Signal Prediction Method Based on ARIMA Model and DWT

Predicting future ECG signal from previous ECG is a difficult task because of its inherent random feature. In order to improve the accuracy of prediction, this paper proposes a novel ECG signal prediction method based on Autoregressive Integrated Moving Average (ARIMA) model and discrete wavelet transform (DWT). Firstly, ECG signals are preprocessed by smoothing. Secondly, ECG signals are decomposed into a low-frequency approximate component and some high-frequency detail components by DWT. Thirdly, ARIMA model and Autoregressive Moving Average (ARMA) model are used to model and analyze the approximate component and the detail components respectively. Finally, the proposed prediction method is evaluated by ECG data from MIT-BIH database. The experiments show that the prediction accuracy of the proposed method is satisfactory, and the RMSE (Root Mean Square Error) and MAE (mean absolute error) of this paper are very small, only 10
^-4
magnitude. Compared with some other ECG prediction methods, the accuracy of the prediction methods proposed in this paper is significantly improved.