Analysis of Baseline Impact on Differential Doppler Positioning and Performance Improvement Method for LEO Opportunistic Navigation

Abstract

The positioning system based on the signals of opportunity (SOPs) of low-Earth-orbit (LEO) satellites can meet the positioning requirements of the receiver in the global navigation satellite system (GNSS) denial environment. However, the received signal does not contain or cannot extract the orbit information and signal emission time of the radiation source, which has become the main error source in this positioning system. Taking the Iridium satellites as an example, a Doppler differential (DD)-LEO (DD-LEO) positioning framework is proposed in this article, and the baseline impact analysis of the differential Doppler positioning system (DDPS) and a performance improvement method are carried out. First, the workflow between a base and a rover is introduced, and the error elimination of the DDPS is analyzed. Second, the equivalent Doppler error models formed by the orbit error at the base and the rover are established, respectively, and the relationship between the baseline and the mean or variance of equivalent Doppler residual in the DDPS is derived. Then, aiming at the problem of poor positioning accuracy due to the inconsistency of space–time reference in the long-baseline DDPS, a signal emission time estimation (SETE) algorithm based on maximum likelihood estimation (MLE) is proposed to further improve the positioning accuracy. Finally, a simulation experiment of the influence of baseline length and direction on the DDPS is implemented, and several groups of field differential experiments with different baseline lengths are completed. The experimental results of long-baseline positioning show that the DDPS can reduce the 3-D positioning error by 48.0% compared with the single station without differential positioning. At the same time, the positioning accuracy of the rover station is further enhanced by 86.3% when the base station adopts the SETE algorithm, which verifies the effectiveness of the proposed algorithm.