A theoretical approach for the dynamic reconfiguration of an on-board antenna pattern and its performance assessment

In TLC satellite systems operating in the Ka-band, the dynamic control of the on-board antenna pattern is recognized as an effective countermeasure against the atmospheric attenuations due to rain, water vapour, etc. The on-board available power is better exploited when distributed on the service area by privileging the sub-regions requiring, at a given time, more power to counteract the local attenuation. Based on this principle the power flux distribution on the Earth surface must be tuned according to weather variations. An efficient algorithm has been devised at the Politecnico di Milano to evaluate the best distribution of radiated power by processing standard weather data from the ECMWF and Meteosat databases. The system performance is related to the random distribution of the meteorological impairments (both in time and in space), and it can only be described in statistical terms. The optimization process consists in determining, at a given time, a set of optimum excitation coefficients for a 64 feeds multiport antenna, which minimize the number of non served users in the service area. Our method is based on an analytical procedure which is less demanding in terms of computational effort with respect to the standard optimization routines based on iterative algorithms. By applying the optimization algorithm to one week of weather pictures provided by the above- mentioned databases (6 pictures per day) the advantage of the adaptivity method can be appreciated and the system effectiveness demonstrated as well.