Simple Explanation (ELI5)

Think of correcting the compass of a swarm of drones flying in formation: if you don’t remove each antenna’s factory quirks, every drone thinks it’s in the wrong spot. With just 3-4 well-placed spots, you can calibrate the whole swarm without checking each drone individually.

Key Advantages & Current Limits

🟢 Pros: With only 3-4 stations and a week of data, sub-centimeter accuracy is achievable, cutting infrastructure costs. 🟢 Pros: The method separates PCO and delays, using a global batch first and then PPP to better isolate each error source. 🔴 Cons: The vertical PCO component is highly sensitive to orbit errors (>10 cm) and clock errors (>5 ns), which can significantly degrade the estimate. 🔴 Cons: Adding more satellites yields diminishing returns beyond ~50, indicating a practical limit set by observation geometry.

Future Outlook & Takeaways

In the next 3-5 years, small but strategically distributed ground networks will automatically calibrate LEO constellations, enabling centimeter-level navigation even in remote or maritime areas. This will make ultra-precise positioning services widely available at low cost, benefiting drones, autonomous vehicles, and geological surveying.

How are LEO antenna errors measured with few ground stations?

Article

100%

Low‑Earth‑orbit (LEO) satellite constellations are reshaping navigation, but two hidden parameters must be known precisely: the antenna phase‑center offset (PCO) and the equipment delay. If mis‑estimated, they can shift calculated positions by several centimeters. Until recently, research focused on one or a few satellites, ignoring the impact of constellation size, orbital altitude, or ground‑station layout.

A Chinese research team devised a method that uses only three to four regional stations, capable of tracking both GNSS and LEO signals simultaneously. "The method combines a batch least‑squares estimate for the PCO and initial delays, then refines the delays with a PPP approach." First, all PCOs are solved in a global batch, then these values are fixed while equipment delays are fine‑tuned using precise point positioning.

Simulations reveal: 1) the vertical PCO component is most sensitive to orbit and clock errors; 2) adding more satellites improves accuracy, but with diminishing returns; 3) with three to four stations and five to seven days of data, sub‑centimeter PCO errors and equipment‑delay errors under 3 cm are achievable. This shows that even with limited resources, ultra‑precise measurements are possible, paving the way for more reliable, widespread navigation systems.

Bonus

100%

The PCO denotes the offset between the physical antenna point and the ideal signal‑origin point. In a LEO constellation, PCO variations are driven by structural deformations, temperature changes, and vibrations. The Z component (toward Earth's centre) is especially sensitive because altitude changes alter observation geometry.

Equipment delay, on the other hand, is the internal transmitter lag caused by filters, converters, and cabling. It is nearly constant but can shift with temperature and component ageing. The proposed method estimates it using a batch least‑squares approach that minimizes residuals between observed and modeled signals, applying ionospheric and tropospheric corrections. After fixing the PCO, PPP is employed, leveraging precise GNSS reference stations to isolate the instrumental delay.

Simulations varied key parameters: altitude (500‑1200 km), satellite count (30‑120), station number (2‑5), and observation span (2‑10 days). Results show that beyond 50 satellites, accuracy gains plateau, indicating a practical limit set by observation geometry. Moreover, orbit errors above 10 cm or clock errors beyond 5 ns significantly degrade the Z‑PCO estimate, while X‑Y components remain more stable. The study suggests that a small, well‑distributed ground network can still achieve accurate calibration, lowering infrastructure costs for future LEO navigation constellations.