Research on Identification Algorithm Based on PPG Signal and Improved Convolutional Neural Network

In this paper, an improved convolutional neural network model is proposed, which is used for the identification of ECG signals while PPG signals are rarely used. The PPG data from the mimic database is denoised by wavelet transform, and then directly sent to the model for automatic sign extraction and recognition. This method can achieve the highest recognition rate of 99.39% on PPG data in mimic database. This makes the identification based on PPG signal meet the requirements of high reliability. With the rapid development of sensors and wearable devices, this research can also meet the universality of one of the requirements of identification.

[1]  Philip de Chazal,et al.  A Patient-Adapting Heartbeat Classifier Using ECG Morphology and Heartbeat Interval Features , 2006, IEEE Transactions on Biomedical Engineering.

[2]  Willis J. Tompkins,et al.  A Real-Time QRS Detection Algorithm , 1985, IEEE Transactions on Biomedical Engineering.

[3]  Petros Spachos,et al.  Feasibility study of photoplethysmographic signals for biometric identification , 2011, 2011 17th International Conference on Digital Signal Processing (DSP).

[4]  N. Maheswari,et al.  An Enhanced Intrinsic Biometric in Identifying People by Photopleythsmography Signal , 2013 .

[5]  Saurabh Pal,et al.  Photoplethysmogram signal based biometric recognition using linear discriminant classifier , 2016, 2016 2nd International Conference on Control, Instrumentation, Energy & Communication (CIEC).

[6]  Y. Y. Gu,et al.  Photoplethysmographic authentication through fuzzy logic , 2003, IEEE EMBS Asian-Pacific Conference on Biomedical Engineering, 2003..