Efficient Ambiguity Search Technique Using Separated Decision Variables

Abstract

In this paper, we propose new separated decision variables that are derived by directly using the normal distribution of each measurement error and allowing substitution of the chi-square variable of the conventional method. In the derivation of the proposed decision variables, we considered not only the related mathematical model, but also the additional unmodelled properties of GPS measurements. Using the sequential pseudo-moving-average technique, we developed a method that easily obtains the combined results of multiple epochs. To verify our proposed algorithm, we analysed its performance using real data and compared the results with those of the conventional method. Our proposed approach performs better than the conventional approach, and effectively reduces computational effort by approximately 60%. Our results demonstrate that our method achieves a solution that is as reliable as the conventional technique, while reducing the time required to only 15% of that required by the conventional technique.