Route-a

Use the map obtained from the central square antennas as the model in CALIB and solve for all the antenna based gains (i.e. set ANTENNA=0 and UVRAN=0 or some high value say $10K\lambda$ in CALIB). Obviously, this is internally contradicting as one is using a point source model for a source which is actually extended (in this case for UVRAN$>2$K$\lambda$). This limitation is indeed reflected in the extremely poor solutions obtained for the antenna gains in this case. Bitten by this, we instead move towards following two more conservative routes.