A technique for estimating the boresight direction of a Global Positioning System (GPS) receiver antenna involves utilization of the relationship between the strengths of received signals and the direction-dependent antenna gain pattern. The technique is fundamentally different from, and much less precise than, other attitude-determination techniques based on interferometry with multiple antennas. The major advantage of this technique is that it quickly gives a coarse estimate, using data from only one antenna. The coarse estimate is not suitable for fine-attitude applications like aiming a telescope or a laser beam, but it can be used, for example, to guide the orientation of a broad-beam communication antenna, to aim a solar panel, or to initialize a fine attitude-determination algorithm or instrument.

The technique is most easily practiced in the case of an antenna with a broad radiation pattern in which the gain decreases monotonically with increasing angle off boresight. The GPS receiver used in this technique must be one that generates data on the signal-to-noise ratio (SNR) of the signal received from each GPS satellite that it tracks. Once the GPS receiver has computed its position from the received GPS signals, the direction to each tracked GPS satellite is known as a byproduct.

The SNR of the signal received from each tracked GPS satellite is taken as a crude measure of the relative strength of the signal and, as such, is used as a weighting value to obtain a vector sum: The unit vector in the known direction to each tracked satellite is multiplied by the SNR for that satellite. The sum of such scalar·vector products for all the tracked satellites is a vector, the direction of which is taken to be the estimated antenna boresight direction. The length of the vector also constitutes ancillary information about the geometric properties of the constellation of tracked GPS satellites.

If only one GPS satellite is being tracked, then the estimated boresight points directly at that satellite; such an estimate is usually erroneous, but it could be helpful in finding other satellites to track and thus obtain a better estimate. When six to eight GPS satellites are being tracked, the estimated boresight differs from the actual boresight by no more than about 15°.

This work was done by Charles Dunn and Courtney Duncan of Caltech for NASA's Jet Propulsion Laboratory. NPO-20323


This Brief includes a Technical Support Package (TSP).
Estimating attitude from GPS measurements on one antenna

(reference NPO20323) is currently available for download from the TSP library.

Don't have an account? Sign up here.