In this work a new optimization technique for the dynamic control of a reconfigurable antenna frontend in TV-sat applications is presented. Reconfigurable antenna is here meant as a radiating system embarked on-board which enables to modify the directivity function over the served region by acting on the excitation coefficients of a set of feeds, so as to counteract at best the attenuation effects of transmission in Ka-Band. In this paper we present, some preliminary results of a study conducted by the Politecnico di Milano on behalf of the ESA and ASI [1]. In the study we assume that a parabolic antenna is illuminated by a set of 64 feeds positioned in the focal plane. The radiation is controlled by beam-forming network containing by a set of Multi-Port Amplifiers (MPA) excited by a set of optimized excitation coefficients. These are obtained by means of an optimization process which takes as input a set of meteorological data (Meteosat and ECMWF images). The service area has been sampled with a grid of about 2000 points, called “pixels”, while the figure of merit (to be minimized) is the total Number of Non Served users (NNS) across the served area, i.e. the sum of users not reached by a sufficient power flux. To this task each pixel is associated to a characteristic curve giving the Number of Non Served Users contained in the generic i-th pixel (NNSi) as a function of the power flux mi impinging on that pixel. The optimization technique is based on an algorithm based on a principle called “derivative slicing principle” which consists in equalling all partial derivatives of the NNSi with respect to mi. This rule leads to k-1 (k = 2000) equations which, jointly with the constraint Σ = i m available power,allow to determine all the excitation coefficients ensuring a “mathematical” NNS minimum conditioned to the given amount of the available power. A significant example of the advantage obtainable with this procedure is presented in the paper. [1] ESA/ESTEC Contract. N.222116/09/NL/ST

`http://hdl.handle.net/11311/581809`

Titolo: | A theoretical limit for the optimum adaptive distribution of the power flux on Earth in TV-sat applications at 20 GHz |

Autori interni: | RESTEGHINI, LAURA PARABONI, ALDO |

Data di pubblicazione: | 2010 |

Abstract: | In this work a new optimization technique for the dynamic control of a reconfigurable antenna frontend in TV-sat applications is presented. Reconfigurable antenna is here meant as a radiating system embarked on-board which enables to modify the directivity function over the served region by acting on the excitation coefficients of a set of feeds, so as to counteract at best the attenuation effects of transmission in Ka-Band. In this paper we present, some preliminary results of a study conducted by the Politecnico di Milano on behalf of the ESA and ASI [1]. In the study we assume that a parabolic antenna is illuminated by a set of 64 feeds positioned in the focal plane. The radiation is controlled by beam-forming network containing by a set of Multi-Port Amplifiers (MPA) excited by a set of optimized excitation coefficients. These are obtained by means of an optimization process which takes as input a set of meteorological data (Meteosat and ECMWF images). The service area has been sampled with a grid of about 2000 points, called “pixels”, while the figure of merit (to be minimized) is the total Number of Non Served users (NNS) across the served area, i.e. the sum of users not reached by a sufficient power flux. To this task each pixel is associated to a characteristic curve giving the Number of Non Served Users contained in the generic i-th pixel (NNSi) as a function of the power flux mi impinging on that pixel. The optimization technique is based on an algorithm based on a principle called “derivative slicing principle” which consists in equalling all partial derivatives of the NNSi with respect to mi. This rule leads to k-1 (k = 2000) equations which, jointly with the constraint Σ = i m available power,allow to determine all the excitation coefficients ensuring a “mathematical” NNS minimum conditioned to the given amount of the available power. A significant example of the advantage obtainable with this procedure is presented in the paper. [1] ESA/ESTEC Contract. N.222116/09/NL/ST |

Handle: | http://hdl.handle.net/11311/581809 |

Appare nelle tipologie: | 04.1 Contributo in Atti di convegno |