A SAR antenna calibration method

Corner reflector targets have been used to calibrate ERS-1 and ERS-2's SAR antenna. The analysis method permits resolution of the sources of variation in reflector backscatter into components by considering the different dimensions in the data. This results in the antenna gain pattern; temporal calibration constant series; calibration target differences, and a pure error term. The pure error term may be further reduced into fluctuations from targets, clutter effects, and spatial variations. The regressional model adopted allows the calculation of a standard deviation in the antenna gain pattern. This was found to be of the order of 0.1 dB, t- and /spl chi//sup 2/-statistics have been devised to compare the antenna gain pattern with ESA's pattern derived from homogeneous distributed targets. Analysis during ERS-1's geodetic phase implied that there were significant differences between day time and night time antenna gain patterns. A rms difference of 0.17 dB was found. This result has not been confirmed during ERS-2's commissioning phase and a possible explanation has been developed.