Abstract: Beamforming in a satellite communications network includes determining a first delay imparted to signals redirected by a first reflecting dish in a satellite, the first delay being caused by motion of the first reflecting dish relative to the satellite. A second delay imparted to signals redirected by a second reflecting dish in the satellite is determined, the second delay being caused by motion of the second reflecting dish relative to the satellite. Calculating, using the determined first and second delays, beamforming coefficients that enable the generation of a single beam by combining a group of signals that includes a first signal redirected by the first reflecting dish and a second signal redirected by the second reflecting dish.

Abstract: Gain measurements of one or more calibration signals communicated by one or more calibration stations and retransmitted by one or more satellite elements may be obtained. One or more weights indicating confidence in the gain measurements for the one or more satellite elements may be computed. A geodetic radiation gain pattern of a satellite element that is a function of a roll, a pitch, and a yaw of a satellite may be accessed. Partial derivatives of the geodetic radiation pattern may be computed with respect to roll, pitch and yaw at one or more points in the first geodetic radiation pattern corresponding to one or more geographic locations. A cost function may be generated and one or more unknown variables may be computed by reducing the cost function. A satellite pointing error may be determined using the computed variable values and the pointing error may be stored.

Abstract: In one implementation, a first measurement of gain of a first calibration signal retransmitted from a first satellite element and a second measurement of gain of a second calibration signal retransmitted from a second satellite element may be obtained. A first expected gain for the first element and a second expected gain for the second element may be accessed. A weight indicative of a confidence in the first measurement may be computed using a gradient of a geodetic radiation pattern of the first element. A weight indicative of a confidence in the second measurement also may be computed using a gradient of a radiation pattern of the second element. A cost function may be generated and values for one or more unknown variables may be computed by reducing the cost function with respect to the one or more unknown variables. The computed values for the unknown variables may be stored.