Development of an algorithm for preprocessing and filtering multimodal biometric data for a software and hardware complex with intelligent control
Keywords:
rehabilitation glove, multimodal data, signal preprocessing, Kalman filter, wavelet transform, motion artifacts, electromyography, adaptive filteringAbstract
A complex preprocessing algorithm is proposed, including adaptive calibration of sensors,compensation of motion artifacts based on the Kalman filter and wavelet transformation, as well assynchronization of heterogeneous signals (electromyography, pressure, temperature, pulse). It is proposed to create a universal system for pre-processing biometric signals, including cleaning initial datafrom noise, correction of distortions and normalization of indicators. This will significantly improvethe quality of the source data before they are further processed by machine learning algorithms.A model for assessing signal quality in real time was developed. doi 10.54708/19926502_2025_29411065Downloads
Published
2025-25-12
Issue
Section
******************************