Minimizing the integer ambiguity search space for RTK

Differential GPS carrier phase measurements have much lower noise and multipath error than that of pseudorange measurements.The result is centimeter accuracy for Real-Time Kinematic (RTK).However, the measurement of the carrier phase has a constant unknown integer ambiguity. Several technical issues are related to solving the integer ambiguity correctly. They are: proper stochastic model, search space definition and initialization, search space reduction, state and standard deviation calculation, validation and rejection criteria for the unique and correct candidate. Search space reduction is critically important. It not only affects the ambiguity resolution speed, but also defines the ambiguity resolution success rate. The smaller the search space, the easier it is to find the unique and correct candidate set. The paper analyzes the integer ambiguity search space in its residual domain.The search space is minimized by: Analyzing the maximum independent integer ambiguity measurement set theoretically; Selecting the best initial measurement set that minimizes the search range of each satellite in the set and reduces the error effects from noise that may cause the wrong integer ambiguity solution for the remaining satellites not contained in the initial measurement set.Since the Residual Sensitive Matrix (S-Matrix) relates the integer ambiguity candidate set directly to post-fix residuals, it is not necessary to compute a fix for each candidate set thus making the integ erambiguity search process much more efficient. Also, minimizing the search space in the residual domain improves the search efficiency significantly and at the same time improves its reliability. Performance issues, such as recursive and weighted search techniques as well as methods for improving reliability, are also discussed in the article.