Abstract

A new technique for ambiguity resolution at long distances is described. It uses the code and carrier measurements on three frequencies in an unusual way. Specifically, it uses an averaging method to arrive at an accurate ambiguity-resolved and refraction-corrected measurement that largely overcomes the disadvantage of the close spacing between the L2 and the L5 frequencies. Of course, it works better when the second and third frequencies are farther apart, such as the Galileo L5 and E6 frequencies. The technique is unique in that there is no requirement to resolve the ambiguities of the fundamental L1, L2 and L5 carrier phase measurements. Instead a wide-lane, but noisy, refraction-corrected carrier phase measurement is formed from two of the three widelane carrier phase differences formed from the difference of pairs of the fundamental phase measurements. These differences are ambiguity resolved using ionosphericmatching code measurements and are then combined into a refraction-corrected composite measurement. While this wide-lane composite is quite noisy, it can be smoothed with a refraction-corrected, composite measurement with much lower noise.

The ambiguities of this low-noise composite measurement are not required since it is simply used to smooth the noise in the wide-lane refraction-corrected composite. By not requiring the stepping from one ambiguity-resolved carrier phase measurement to another, it is largely immune to clock differences at the different frequencies which can sabotage the stepped approach. In addition, because the initial ambiguity resolution is done with wide-lane combinations, the reliability of the ambiguity resolution is robust and relatively insensitive to the presence of small code-carrier biases.

The geometry-free approach of individually resolving the ambiguities removes the tropospheric refraction from the ambiguity resolution problem. Thus, the final smoothed, refraction-corrected composite measurement is insensitive to both ionospheric and tropospheric refraction effects. Though the smoothing process may require some minutes to reach the optimal accuracy level, the result should significantly extend the ranges over which RTK results can be obtained without requiring the modeling of the ionosphere.