Statistical Assessment of New Methods to Optimize the Power Flux Distribution in TV-sat Broadcasting at 20 GHz

In this paper we present the results of a statistical performance assessment of a new advanced TV-sat system operating in the ka band. The on-board transmission system is based on the idea of reconfigurable antenna, i.e. the antenna directivity can be modified so as to irradiate more power to the regions affected by bad weather. The core is a power flux distribution optimizer based on the “derivative slicing” method, whose mathematical basis is described in the companion paper [1] presented at this Conference. Here, the above technique is assessed by verifying the system performance through a statistical analysis over six-months. The simulations make use of a meteorological database provided by the European Centre for Medium-range Weather Forecasts (ECMWF). The constraints unavoidably posed by the usage of a physically realizable Beam-Forming Network (BFN) are taken into account as well. The results presented allow a preliminary assessment of the predicted advantage (in terms of availability) of an adaptive system over non-adaptive systems currently implemented, which are based on long-term climatological data. The BFN used to distribute the signal power across the service area (Europe) is constituted by a set of 8x8 multiport amplifiers, each driven by a Butler-Like Matrix (BLM), covering the area with 64 beams. The BFN excitation coefficients are refreshed every 3-6 hours, in accordance with the update rate of the input meteorological data. Our analysis shows that the number of users lost due to atmospheric effects is reduced by a factor of about 5 with respect to the case of a non-reconfigurable system fed by the same transmitted power. Morevoer, the theoretical reduction factor achievable by the "derivative slicing” method removing the system constraints (i.e. the BLM elements) would increase to 10.