Variable duration fixed failure rate ambiguity resolution

Abstract

In global navigation satellite system ambiguity resolution, it is common to use a configurable acceptance test, like the popular ratio test, to control the failure rate. Work in the literature proposes acceptance tests which let users directly choose an acceptable failure rate, but it focuses on instantaneous and fixed duration ambiguity resolution (where results may be success, failure, or undecided). We present a fixed failure rate acceptance test which avoids undecided results by varying the duration of the ambiguity resolution window. The proposed method uses Monte Carlo simulation with a time-correlated noise model to select acceptance test parameters that will minimize mean ambiguity resolution time while maintaining the target failure rate. Results of real-world experiments with single-frequency data from receivers with patch antennas show with 24 h of stationary data the method produces a mean ambiguity resolution time reduction of 52 % (from 525 to 250 s) compared with a ratio test. In a test with 90 min of kinematic data, the reduction was 33 % (from 620 to 410 s).