Array-Aided GNSS for Precise Determination of Ionospheric and Tropospheric Delays With Integer Ambiguity Resolution

Ground-based global navigation satellite system (GNSS) observations are essential for acquiring high-precision ionospheric and tropospheric information due to their superior temporal and spatial resolution, making them invaluable in space weather monitoring and meteorology. However, the widely used single-station precise point positioning (PPP) technology faces accuracy challenges due to the need to estimate a large number of parameters and the presence of float ambiguities. Array-aided GNSS technology holds promise for enhancing atmospheric information extraction through redundant observations. Yet, traditional array-aided models struggle to extract tropospheric delays, as tropospheric parameters are absorbed by satellite clocks during the elimination of rank deficiencies. In this article, we present an array-aided precise atmospheric delay determination (A-PADD) method that incorporates integer ambiguity resolution (IAR). Building on the PPP framework, we develop a full-rank model that allows for the simultaneous estimation of ionospheric and tropospheric parameters, utilizing a short baseline for rapid IAR. Through canonical decomposition (CD) theory, we analytically demonstrate that while IAR does not enhance ionospheric delay estimation, it improves the determination of tropospheric delays. Empirical data corroborate this theoretical insight. Our experimental results show that the A-PADD method surpasses the traditional PPP method in terms of accuracy and stability of atmospheric delay extraction. Specifically, for ionospheric delay determination, A-PADD accelerates the initialization process through additional redundant observations. For tropospheric delay determination, A-PADD improves the extraction accuracy by 23% compared to traditional PPP and shows greater consistency with the international GNSS service (IGS) final product. Therefore, within the array-aided framework, for applications solely focused on ionospheric delay estimation, higher accuracy can be achieved without IAR, thereby reducing computational requirements. For applications requiring both ionospheric and tropospheric delay information, implementing IAR is advantageous, particularly during the initialization phase of tropospheric delay determination.