Base station antenna near-field radiation pattern distortion analysis

Abstract

When estimating the potential radiation hazard of a base station antenna system, one often relies on the antenna far-field radiation pattern, while the estimation often refers to the antenna near-field zone. It is widely accepted that this leads to overestimation of the field amplitudes. Some recent works suggest the use of modified analytical models to avoid the overestimation problem. However, all of these references refer only to a single antenna in the free space. This work shows that potential near-field radiation pattern distortion due to conductive objects in the close proximity of the antenna (like in multiple- antenna configuration) makes many models inapplicable. In that case, in the directions outside the mainlobe of the antenna, sidelobes are shifted and changed in an unpredictable manner. The nulls are either shifted, or filled so they do not really exist. The mainlobe itself can get an additional gain of a few dB in unpredictable directions. This suggests that even the use of the far-field pattern can sometimes lead to underestimation of the field. Considering that many real antennas have some conductive object nearby, especially in multiple antenna configurations, this work suggests that for truly conservative, worst-case electromagnetic field estimation, far-field pattern should be used for the mainlobe, with the uncertainty of a few dB. For all other directions, the protective envelope should be used as the radiation pattern, with the gain in all directions equal to the highest sidelobe gain. No nulls should be considered to exist in the near field radiation pattern.