| Autor: |
Gu, Xiaobo, Qiu, Zeyang, Wang, Yanjiao, Jiang, Wei |
| Zdroj: |
GPS Solutions; Jul2024, Vol. 28 Issue 3, p1-16, 16p |
| Abstrakt: |
The inter-satellite link (ISL) has been received increasing attention, as it is a potential way to achieve autonomous clock synchronization for envisioned space-based satellite networks with minimal ground segment capability. Existing satellite clock synchronization solutions either rely on prior information or modelling the relative motion by an approximated polynomial. In this paper, we propose a deep learning approach based on long short-term memory (LSTM) to decouple the clock parameters from pseudo-range measurements. The process of clock parameter estimation solely relies on the observed pseudo-range measurements, and the prior information of position and velocity are not required and the nonlinear relative motion process is modelled by training on historical data. The simulation results show that the proposed method outperforms the benchmark solutions in terms of accuracy. [ABSTRACT FROM AUTHOR] |
| Databáze: |
Complementary Index |
| Externí odkaz: |
|
Full text from SpringerLink