Research Article
Individual Identification Based on ECG Signal Driven by Multi-layer LSTM and EEMD Algorithm
@INPROCEEDINGS{10.1007/978-3-031-23902-1_17, author={Xueting Wang and Xiaohong Zhang}, title={Individual Identification Based on ECG Signal Driven by Multi-layer LSTM and EEMD Algorithm}, proceedings={Mobile Multimedia Communications. 15th EAI International Conference, MobiMedia 2022, Virtual Event, July 22-24, 2022, Proceedings}, proceedings_a={MOBIMEDIA}, year={2023}, month={2}, keywords={ECG biometrics Human identification Long short-term memory Ensemble empirical mode decomposition}, doi={10.1007/978-3-031-23902-1_17} }- Xueting Wang
Xiaohong Zhang
Year: 2023
Individual Identification Based on ECG Signal Driven by Multi-layer LSTM and EEMD Algorithm
MOBIMEDIA
Springer
DOI: 10.1007/978-3-031-23902-1_17
Abstract
As people attach great importance to the field of information security, identification technology based on biometrics has been widely developed and applied. However, biometric identification technology based on face and fingerprint has the disadvantage of weak anti-counterfeiting and easy to prevent. Electrocardiogram (ECG) signals have high anti-counterfeiting properties of living body recognition, which makes the identification technology based on ECG signals have great development potential in the field of information security. This paper proposes an ECG identification algorithm based on Ensemble Empirical Mode Decomposition (EEMD) and Long Short-Term Memory (LSTM). First, the one-dimensional non-stationary and nonlinear ECG signals are decomposed by EEMD, and the Intrinsic Mode Functions (IMFs) of each layer are extracted in the time-frequency domain. The vector is used as the input layer of the multi-layer LSTM to complete the feature classification and output the individual identification result. The recognition accuracy of the proposed model is 95.47% (ECG-ID datasets) and 96.74% (Physionet/Cinc Challenge 2011 datasets), indicating that the proposed model can achieve a high recognition accuracy and capacity for generalization.
