Biometric data prediction based on deep learning and piezoelectric sensors

Abstract

Biometric data prediction is an important research direction of portable sleep monitoring devices such as piezoelectric ceramics. The current sleep monitoring sensors have limited anti-interference ability and accuracy, which limits the accuracy of prediction results. In order to improve the accuracy of sign data measurement of piezoelectric sensors, this paper proposes a depth measurement method based on LSTM based on the analysis of existing prediction models. The experimental results are compared with other time series prediction models. The results show that the predicted value of the LSTM model is close to the real value, and the mean absolute error and root mean square error of the LSTM model are lower than those of RNN and ARIMA models, which verifies the prediction performance of the LSTM model and can provide effective sign data information for sleep monitoring equipment.rs, Footnotes, or Math in Paper Title or Abstract.