TL;DR
This paper introduces a new Ukrainian-Chinese network of Doppler stations designed to precisely track Low Earth Orbit (LEO) satellites. By leveraging a cost-effective system and advanced signal processing, the network successfully refines satellite orbital parameters, demonstrating a significant reduction in tracking errors.
What Problem Does This Paper Address?
Accurately tracking LEO satellites is crucial for a variety of applications, from space debris monitoring to scientific observation. Traditional methods, like optical tracking, face challenges due to the satellites' rapid movement and limited visibility windows (dawn/dusk). While radio tracking offers advantages like independence from atmospheric conditions and real-time operation, achieving high accuracy in refining orbital parameters requires robust and synchronized data. This paper addresses the need for a more accessible, accurate, and reliable system for LEO satellite orbit refinement, particularly by establishing a collaborative network between Ukrainian and Chinese research institutions.
Methods
The core of this research lies in the development and deployment of a network of Doppler stations. These stations utilize a passive radar approach, measuring the Doppler frequency shift of radio signals emitted by LEO satellites in the 430-440 MHz UHF band. The stations are built using readily available "DVB-T+DAB+FM" receivers and an eight-section omnidirectional antenna system. A key innovation is the automated antenna selection, which dynamically chooses the optimal antenna section based on pre-calculated satellite trajectory data from the NORAD catalog. This ensures precise tracking even with the satellite's rapid movement.
For data processing, the paper describes a two-part software system:
- Real-time operational software: This component handles data acquisition, signal processing, and synchronization with GPS time. It includes custom software (SatTrack, TSIP_SYN) and the open-source HDSDR program for receiver control.
- Post-processing software: This is where the orbital refinement takes place. The WinMNK software, based on the Least Squares Optimization (LSO) method, is used to refine orbital elements using the collected Doppler shift data. It employs both numerical (SGP4/SDP4) and analytical motion models.
Datasets
The primary dataset used for evaluation consists of Doppler measurements from the OSCAR-19 satellite. These observations were collected by the Ukrainian-Chinese network of Doppler stations on September 21-22, 2023. The paper also references the use of NORAD Two-Line Element (TLE) data for initial satellite tracking and orbital element refinement. No new benchmarks or datasets were introduced in this paper; rather, existing data and established catalogs were utilized.
Key Insights and Ideas
- Cost-effective LEO tracking: The use of "DVB-T+DAB+FM" receivers and an automated antenna system demonstrates that high-accuracy LEO satellite tracking can be achieved with relatively low-cost hardware.
- Automated antenna switching: The dynamic selection of antenna sections based on predicted satellite paths is a crucial element for maintaining continuous and precise signal reception, especially for fast-moving LEO satellites.
- Networked approach enhances accuracy: Synchronous observations from multiple stations (Mykolaiv and Sheshan) help mitigate systematic errors arising from individual station characteristics and geographic location.
- Doppler shift as a powerful refinement tool: The study confirms that Doppler frequency shift measurements, when processed correctly, provide valuable data for significantly improving the accuracy of LEO satellite orbital parameters.
Key Results
- The Ukrainian-Chinese Doppler station network was successfully created and tested.
- Five successful identifications of the OSCAR-19 satellite were conducted using its NORAD catalog orbit.
- Four refined sets of orbital elements for OSCAR-19 were obtained.
- Refining orbital elements using sequential passes resulted in a decrease in both systematic and random errors of radial velocity differences. Specifically, Table 4 shows that for the RI "MAO" station, the mean radial velocity difference decreased from 0.15 m/s to 0.08 m/s, and the standard deviation dropped from 4.31 m/s to 2.48 m/s. Similar improvements were observed for the SHAO station.
- Figure 5 and 6 illustrate the dependence of radial velocity differences on signal amplitude, showing a reduction in discrepancies after orbital refinement.
Limitations and Future Work
Limitations: * The paper notes that standard spectral analysis methods can become ineffective for high-frequency change rates, leading to errors in frequency determination. * While the system is described as passive, the paper mentions that other Doppler systems (like ARGOS) and active radar systems (like GRAVES) exist, implying that this passive approach might have its own limitations in certain scenarios (though not explicitly detailed as a limitation of this system).
Future Work: The authors propose several upgrades to further enhance the network's performance and precision:
- Increasing reference oscillator stability: Upgrading the "DVB-T+DAB+FM" receiver's reference oscillator to a high-precision frequency synthesizer (from 10⁻⁶ to 10⁻⁹).
- Improving time synchronization: Integrating a 1 PPS (pulse per second) signal into the frequency reference field for more precise synchronization of measurement data with absolute time. These improvements are expected to reduce synchronization errors and boost overall Doppler measurement accuracy.
Summary updated