Hacker News Reader

Comment by causi | Hacker News Reader

Interesting that whatever technology underlies the SBAS system isn't just integrated into GPS. It would be nice to be able to map property lines with just my phone.

View on HN

The US government does provide an SBAS called the WAAS, operated by the FAA. It's also had a hand in designing and operating a number of other augmentation systems [1], including CORS, GDGPS, and NDGPS. As an aside, don't you just love how many government acronyms there are?

[1] https://www.gps.gov/systems/augmentations

WAAS doesn't cover Australia.

Wasn't saying it did. MSAS (Japan) can be used in AU if you accept the performance degradation, but the lack of a public SBAS for the southern hemisphere is a big reason why SouthPAN is being brought online.

FAA sure loves their TLAs. Some acronyms (or initialisms, for the pedants), like VOR, are TLAs inside TLAs!

It seems that GPS signals can be used for higher accuracy if the receiving equipment uses two frequencies instead of one. This can freely be done by civilians, but requires more expensive equipment. Perhaps this solution with SBAS requires a cheaper setup?

https://www.gps.gov/systems/gps/performance/accuracy/

Multi-frequency GPS is already available in many inexpensive civilian devices such as smart watches and cell phones. It improves accuracy a little, primarily when the device doesn't have a clear view of the sky like when next to buildings. But it's still not as precise as SBAS.

The highest precision GPS receivers require dual-frequency reception and perform "carrier differential GPS" - any time people are talking about GPS with 2cm accuracy, that's what they're talking about.

The GPS signal has a digital signal (with a wavelength of ~300m) modulated with a sine wave (with a wavelength of ~20cm) - you can measure the 20cm signal much more precisely than the 300m signal, but it's 'ambiguous' because it repeats every 20cm and you don't know how many waves of the sine have passed.

If you can receive two frequencies, giving you two slightly different sine waves, you can multiply them together and get a sine wave with a longer period, which (together with some clever maths) lets you bridge the gap between the unambiguous-but-imprecise digital signal and the precise-but-ambiguous sine wave.

You also need a fixed base station at a known location - or access to a commercial network of such stations.

You can use the cell towers to assist in mapping it correctly too as they are fixed position and the location is well known. Pretty sure the QCOM patents on it are out of patent by now. Not sure if the android/ios api will expose the whole list of towers the chip can see though. The chip can do it though.

There are GPS RTK systems designed for use by surveyors. They can cost over $10k, but you can also get much cheaper kits that you can put together yourself from SparkFun -- like less than $1k, easy.

Typically these systems are broadcasting data obtained from many private regional reference stations on the ground. Look up “VRS RTK” as an example.

We don’t want other countries weapons guided by our gps satellites!

This isn't 1993 anymore. Beidou and Glonass (and Galileo) are global networks that do the same things for those other countries, and the ITAR restrictions on GPS receiver altitude and velocity are increasingly able to be overcome by a hobbyist in a garage, much less a nation-state.

It doesn't matter - the speed of light in air varies by enough that no satellite can provide the needed accuracy alone.

Not quite. Geodetic GNSS systems use two different carrier frequencies to measure and correct for atmospheric dispersion. This reduces the error to sub-mm levels. The inaccuracy of real-time GNSS is due to unknown variations of the satellite orbits, satellite clocks, and earth rotation.

Reply on news.ycombinator.com