Anthropogenic noise in observations of natural electromagnetic field

   Observations of geophysical electromagnetic fields always contain noise,
   besides the meaningful data; this noise mainly has anthropogenic origin.
   Sometimes this noise is an insuperable obstacle for further processing.
   Thus, noise analysis is a necessary part of experimental data correct
   processing and interpretation.

   The paper analyzes anthropogenic noise in the registration of the Earth
   natural electromagnetic field. It was discovered that this noise has
   coherent and incoherent components. A model for noise was developed and an
   algorithm and its software realization to eliminate the most significant
   noise components with the longest coherence time were proposed. Tests of
   the algorithm on real data showed quite good efficiency.