| An important property of AAs is the varying noise coupling between the antenna elements, and consequently varying system noise temperature, as a function of scan angle, described by the active reflection coefficient. For Apertif, used as an aperture array (AJDI of August 6 and 27, 2010), a maximum noise temperature contribution to the array system temperature of 13 K is calculated at 45° scan angle, 25 % of the system noise budget. Another noise contribution, at large scan angles, comes from noise pick-up from the environment near the horizon and depends on the beam width of the scanning beam. The accompanying figure shows measured noise temperatures as a function of scan angle for arrays of 16 and 49 active elements of the Apertif tile at 1400 MHz, compared to simulation results for both arrays. The detailed effects of the location and height of trees and buildings are not taken into account in these simulations and explain the bigger differences between measurements and simulations at large scan angles.
It is interesting to see how the measured array noise temperature varies with scan angle and beam steering in general. The figure shows the variation in noise temperature for a maximum scan angle (from broadside) of 85°. Up to scan angles of 40° the noise temperatures only slowly vary due to the changing noise coupling, quite well in agreement with simulations. For larger scan angles the broad beams introduce noise from the environment, which dominates over the calculated variation in noise coupling. Obviously the narrower 49-element beam allows further scanning without increase in noise than the 16-element beam. It is also shown that the narrower beam more effectively samples the environmental temperature near the horizon, giving a higher system noise temperature.