For the purpose of human protection from possible harmful effects, but also for the purpose of continuous monitoring of field levels, calculation and measurement of electromagnetic fields in the vicinity of base stations are required. Apart from reference levels (electric and magnetic field intensity and power density) it is convenient to consider dosimetric quantities such as specific absorption rate and transmitted power density for the purpose of the assessment of exposure levels. This thesis deals with measurement methods used for the field levels in the vicinity of base stations. Calculation of field levels is carried out using multiple models taking into account not only direct, but also the reflected wave component, respectively, from the two-layer conducting half-space. In addition to deterministic cases, stochastic analysis is undertaken by performing uncertainty quantification of input variables necessary for the field calculations. Mean and standard deviation are determined by using Monte Carlo and Stochastic Collocation, while analysis of variance is used to determine individual and mutual influence of input variables to the output of interest, i.e. electric field intensity. In addition to the consideration of the external fields, a model for the assessment of the internal field is developed aiming to determine specific absorption rate and transmitted power density, respectively, within the two-layer tissue model. Within the framework of stochastic model of the internal field, uncertainty quantification of the input parameters is carried out via stochastic collocation for variables describing the tissue properties (specific conductivity, relative permittivity and tissue layer thickness) and sensitivity analysis, as well.