Efficient diagnosis of radiotelescopes from far-field data

The paper presents an innovative method for the diagnosis of reflector antennas in radio astronomical applications, which optimizes the number and the location of the far field sampling points exploited to retrieve the antenna status in terms of feed misalignments. In this way the measurement time length process is drastically reduced to minimize the effects of the time variations of the measurement setup, as well as the idle time forced by the maintenance activity. The effects of the feed misalignment are modeled in terms of an aberration function, properly expanded on a set of basis functions in order to preserve the linear relationship between the unknown parameters defining the antenna status and the far field pattern, assumed measured in amplitude and phase. Thanks to the optimization of the Singular Values behavior of the relevant linear operator, in its discrete form, the number and the position of the samples is found. The numerical analysis shows the effectiveness of the method in the simple case of a phase aperture affected by tilt only, even if the approach can be extended also to higher order aberrations. The performances are estimated with a comparison with a standard approach, based on the acquisition of the far field pattern by means of a uniform Cartesian grid defined according the Nyquist criterion and requiring a number of field samples significantly larger.