Modeling of Wet Antenna Attenuation for Precipitation Estimation From Microwave Links

Authors:
Publisher website
Google Scholar
Abstract
This letter presents an analysis of wet antenna attenuation (WAA) from an electrodynamic perspective. The WAA is a relevant part of the path attenuation determined from the received signal level (RSL). In the case of extracting rain rates from path attenuation, as it is done with data from country-wide commercial microwave link (CML) networks, the WAA is a significant error source and leads to overestimation of rainfall. To handle this effect, a model, based on the three antenna parameters, reflectivity, efficiency, and directivity, is developed. The model allows the variation of those three parameters depending on the degree of wetness and shows that WAA is composed of variations of antenna impedance, antenna efficiency, and antenna directivity. The suitability of the theoretical model was assessed by a full-wave numerical simulation of a pyramidal horn antenna at 18 GHz with homogeneous water film on the surface of the antenna cover. The result of this simulation is a quantification of these three subeffects. For verification of the simulation, an antenna, similar to the antenna in the simulation, was manufactured and a controlled sprinkler device to generate a water film was built. The adjustable water flow rate determines the thickness of the water film. The parameters were generated by measuring the near field and the reflectivity of the antenna in an anechoic chamber, with a near-field scanning system and a vector network analyzer, under different flow rates equaling different water film thicknesses. From these results, the reflectivity, directivity, and efficiency were computed and compared with the results of the full-wave simulation. It is found that the reflectivity is linked to water film thickness in a such way that the WAA can be deduced from monitoring the antenna reflectivity.
Funding Information
  • Deutsche Forschungsgemeinschaft (SI 2034/1-2, KU 2090/7-2)

This publication has 18 references indexed in Scilit: