Complexity and Nonlinearity Analysis of the Time Series of Electric Field Intensity

In this work, the conditional entropy (CE) and the method of surrogate data are applied on time series of electric field (EF) intensity to explore their degree of complexity and the possible presence of nonlinear dynamics. The time series were obtained during the wide-band cumulative EF intensity monitoring, performed by one sensor of the Serbian EMF RATEL monitoring system installed in the campus area of the University of Novi Sad, and are re-sampled at one sample per two hours over consecutive time epochs of one month duration. The field intensity measurements during the years 2019 and 2020 allowed us to explore the effects of mobility restrictions related to the COVID-19 pandemic on the patterns of complexity and nonlinearity underlying the EF dynamics. Using the kernel-based estimator of CE, which realizes the well-known Sample Entropy measure, and generating phase-randomized Fourier transform surrogate series, we highlight the presence of predictable but nonlinear EF dynamics associated to months of regular population mobility, while months with reduced human mobility result in more complex but linear EF patterns. In perspective, this analysis may serve to complement standard evaluation of EF intensity and variability, with implications ranging from safety assessment to optimization of sensor placement and social mobility analysis.