Improving the initialization speed for long-range NRTK in network solution mode

Abstract

Initialization speed is one of the most important factors in network real time kinematic (NRTK) performance. Owing to the low correlation among the error sources of reference stations, it is difficult to fix reference station ambiguities of long-range NRTK quickly. In traditional reference stations ambiguity resolution (AR) methods, baselines are usually solved independently which is called baseline solution (BS) mode in this study. Because the correlations among baselines are not taken into consideration in ambiguities estimation, the AR speed is slow. Generally, tens of minutes or longer time is required to initialize. We propose a network solution (NS) mode approach, in which the correlations among the double-difference ambiguities (DDAs) as well as double-difference ionospheric delays (DDIDs) of different baselines are considered in estimating float ambiguity solutions. Experimental results show that the float ambiguity solutions obtained are more accurate with an improved consistency. Thus, initialization speed is significantly increased by 18% in NS mode.