Abstract
This study proposes an original methodology for modeling an outpatient electrocardiographic monitor with a focus on adaptive processing of biomedical signals. A mathematical model of the device has been constructed, incorporating signal parameterization, suppression of noise artifacts, extraction of informative features (RR intervals, heart rate, and heart rate variability indices), and intelligent arrhythmia classification using machine learning algorithms. A distinctive feature of the system is the implementation of a variable sampling rate that automatically adjusts according to signal quality and functional load, thereby optimizing power consumption while maintaining high monitoring accuracy. Computer-based simulations were carried out in the MATLAB/Simulink environment and complemented with experimental validation on both real and synthetic ECG recordings. The obtained results demonstrated reliable detection of QRS complexes (with an accuracy of up to 98.7%) and robust calculation of HRV metrics under noise and artifact distortions. The developed model can serve as a foundation for further optimization of portable ECG monitors, their integration into telemedicine platforms, and the design of intelligent algorithms for early arrhythmia detection.