A Multi-Constraint Co-Optimization LQG Frequency Steering Method for LEO Satellite Oscillators

Area of Science:

• Aerospace Engineering
• Control Systems
• Signal Processing

Background:

• High-precision time-frequency systems are crucial for real-time centimeter-level positioning in low Earth orbit (LEO) satellites.
• Oven-controlled crystal oscillators (OCXOs) used in LEO satellites suffer from long-term stability issues, degrading positioning accuracy due to clock error accumulation.
• Existing systems face challenges in balancing clock bias accuracy, frequency stability, and phase continuity under size, power, and cost constraints.

Purpose of the Study:

• To develop a novel frequency precision steering method for OCXOs in LEO satellites.
• To improve the time-frequency performance of OCXOs, addressing limitations of conventional clock sources.
• To provide a low-cost, high-precision timing-frequency reference solution for LEO satellite navigation.

Main Methods:

• Proposed a linear quadratic Gaussian (LQG) frequency precision steering method integrating a four-dimensional constraint integrated (FDCI) model and hierarchical weight optimization.
• Refined an improved system error model to quantify covariance components (Σ11, Σ22) of the LQG closed-loop control system.
• Utilized a priority-driven collaborative optimization mechanism with the FDCI model to systematically determine weight matrices for robust tradeoff among performance criteria.

Main Results:

• Reduced clock error Root Mean Square (RMS) to 0.14 ns, a 37% improvement over conventional PID control.
• Achieved multi-timescale frequency stability enhancement: 9.38 × 10^-13 at 100 s and 4.22 × 10^-14 at 10,000 s (three orders of magnitude over free-running OCXO).
• Demonstrated 93% stability improvement compared to pure Kalman filtering and contained quantization noise impact within 13% for short-term stability.

Conclusions:

• The proposed LQG frequency precision steering method effectively enhances OCXO time-frequency performance for LEO satellite applications.
• The systematic integration of theoretical constraints and performance optimization offers a robust solution for balancing multi-dimensional requirements.
• This approach provides a cost-effective, high-precision timing-frequency reference, crucial for advanced LEO satellite navigation services.