Static and dynamic evaluation methods for a VDT periphery electromagnetic environment: regression and/or inverse regression models and a dynamic digital filter

摘要

In this paper, a methodological trial of static and dynamic evaluation for the near-electromagnetic-filed leaked from an actually working VDT have been proposed. As a static method, in order to estimate one from another by extracting various interrelation characteristics between sound and electromagnetic waves, from some phenomenon-oriented viewpoint, we have proposed 2 kinds of extended regression and inverse-regression system models reflecting hierarchically not only linear correlation information of the lower order but also nonlinear correlation information of the higher order. As a dynamic method, in order to estimate electromagnetic waves leaked from a specific VDT with an existence of a background electromagnetic noise in the real environment, from an operation-oriented viewpoint, we have proposed a practical estimation algorithm derived by applying the well-known least mean squared method for the moment statistics with several orders. Finally, the validity and effectiveness of these proposed methods are experimentally confirmed too by applying them to the actually observed data.