高精度全球導航衛星系統運用的關鍵在於如何正確及有效率地求解整數相位模稜。 模稜搜尋與解相關技術是解決模稜求解問題的方法之一。傳統上，解相關技術所使用的轉換矩陣其元素的整數限制可確保逆轉換後的候選解仍能保持整數性。但是這限制同樣也使得解相關難以完美。 使用實數轉換矩陣可得到零相關也就是完全對角線化的協方差矩陣。在這新域中的一個空間可用做為門檻，因此這零相關域亦被稱為門檻域。使用本研究所提出之ZETA方法，來自傳統整數轉換所得的候選解數量能夠再次減少。 使用ZETA時，有可能會發生候選解全部遭到剔除而造成無解的情況。本研究使用部分模稜求解處理這情況。部分模稜求解允許將部分的相位模稜做為實數未知進行求解。透過將部分相位模稜求解為實數，候選解將能夠更容易地通過門檻。 實驗顯示本研究所提出之方法能夠在不降低成果精度的情形下，提升演算效率，且模稜求解求解成功率亦能獲得提升。 ;The key point of accurate and precise application of Global Navigation Satellite Systems is how to obtain integer carrier phase ambiguity correctly and efficiently. One of the ways to solve the ambiguity resolution problem is ambiguity searching technique with an ambiguity decorrelation technique. Traditionally, an integer-valued limitation of the transformation matrix of decorrelation technique ensures the integer characteristic of candidates existing after the inverse transformation, but it also makes the decorrelation imperfect. A zero correlation domain or a complete diagonalization covariance matrix could be obtained by the using float transformation matrix. A space in this domain will be used as a threshold, hence the zero correlation domain is called threshold domain. The number of ambiguity candidates based on integer transformation could be reduced through the proposed ZETA method. ZETA might reject all of candidates and make the ambiguity resolution being no solution. In this research, the partial ambiguity resolution is used to cope with this situation. Partial ambiguity resolution allows some of the resolved of ambiguities to be float-valued ones. A candidate will be easier to pass the threshold with some of ambiguities being solved as float solutions. The experiments in this paper prove that the method could make the ambiguity resolution become more efficient without decreasing the accuracy. The success rate could also be improved by proposed method.
