Method of Nonlinear Filtering of Observations from Systems of Vector and Scalar Magnetometers

A method of nonlinear filtering of observations from systems of vector and scalar magnetometers is proposed. The method is based on the calculation of sliding local approximating model functions and weighted averaging. An algorithm intended for the filtering of observations from magnetometer systems based on sliding local approximating piecewise linear model functions (reduction of initial functionals to quadratic forms, calculation of parameters of sliding local models, and implementation of their weighted averaging) has been developed. Test results are presented for the filtering algorithm on model and observatory INTERMAGNET observations with a 1-min discretization of the magnetometer system. The efficiency of the developed filtering algorithm is estimated with statistical modeling.