Analysis of methods for modeling human daily thermometry data

Abstract. Mathematical and computer modeling of daily thermometry allows to study processes human thermal homeostasis more deeply. In practice, data is obtained using a digital thermometer, which autonomously reads the temperature skin in certain time intervals. The aim present work analyse methods processing data. first method consists applying linear discrete stochastic models state space with Gaussian noises known vector input actions, while estimation performed by covariance Kalman filter. second assumes that actions unknown, S. Gillijns B.D. Moor algorithm used process An alternative option use model an extended filtering algorithm. third takes into account presence anomalous measurements (outliers) measurement data, correntropy filter proposed for their effective filtering. Numerical experiments MATLAB were carried out order compare quality algorithms. Modeling was three-dimensional 3dDRCM (3-dimension Discrete-time Real-valued Canonical Model). results can be processes, example, reaction athlete’s body received load.