The technology may have been around for nearly 40 years, but satellite-based navigation systems are still rolling out and still have a lot to offer tomorrow's applications – and it's not just about better precision positioning.
It was early April 2013, and Bradford Parkinson was happily driving along the I5'highway in California. Suddenly, the sizeable semi-trailer riding next to him in the lane began moving erratically from side to side. Parkinson was alarmed to see that it was veering into his lane. He drew up alongside the driver's window. "Sure enough," Parkinson said, "the joker was texting."
This is just the kind of thing that Parkinson wants to avoid by using automated driving technology. He also wants to save commercial airliners landing in the US a vast quantity of fuel each day, and reduce the amount of nitrogen used in farming by 10 per cent, which may go some way towards saving vulnerable marine life in US coastal waters.
If any other 78-year-old engineer presented that list, you might call them a bit crazy – but then Parkinson is one of the fathers (along with Roger Easton and Ivan Getting) of the one technological system that can help achieve all of those things: the US Global Positioning System – or GPS, as it is known around the world that it surveys.
The history of the generic geographic positioning system dates back to 1973, when Parkinson led a meeting at the Pentagon to discuss a system that would eventually populate a constellation of 24 satellites around the world to provide global coverage, with several in reserve. The satellites began launching in 1978, on old Intercontinental Ballistic Missiles. There have been 53'successful launches since then, with 31'satellites still operational.
The satellites send a signal containing precise timing information from onboard atomic clocks. These signals can be used by receivers on the ground, which need information from four satellites to pinpoint their position three-dimensionally. The system is passive, meaning that the satellites generate the data, and don't need information from a receiver to work.
The orbits of the satellites are already known, but the system also relies on ground stations around the globe, operated by the Department of Defense (DoD) and the National Geospatial-Intelligence Agency. This tracking network processes the satellite signals in detail and provides the DoD with error-corrected satellite positions, timing, and Earth-orientation data. It is also broadcast back to the satellites to make their own signals more accurate.
For several years, the US system operated on two frequencies – a civilian and a military one. The civilian frequency was dithered (noise used to randomise quantisation error), with errors deliberately introduced to reduce its native accuracy to 100m. This philosophy, called 'selective availability', was switched off in 2000 under President Clinton, which increased native accuracy to around 10m, in line with accuracy found on the military-usage band.
However, mobile receivers, as used by terrestrial cars and aeroplanes, don't have the precise positioning and timing data used by fixed ground stations. Interference in the ionosphere and slight timing variations is what restricts their accuracy to 10m when using a native GPS signal. The GPS system can be made far more accurate, however, down to a sub-three-metre level using a process known as differential processing.
In differential processing, a fixed reference station on the ground calculates the errors from all the satellites, and then communicates this information to mobile GPS receivers. They can then cross-check the error correction signal against the satellite signals, and use the difference to precisely position themselves. This is the basis for the Wide Area Augmentation System (WAAS), which was deployed by the US to meet the accuracy requirements for aviation set by the Federal Aviation Authority (FAA).
It is the FAA's next-generation air-traffic control system, incidentally, that will make those fuel savings alluded to earlier. Instead of using a stepped descent, where aircraft move down in altitude and then hold, Parkinson says that some airports are already experimenting with tailored arrivals, where GPS is enabling incoming aircraft to make a gradual descent along a continuum.
Specific applications requiring even more accuracy, such as surveying, can use additional technologies, such as carrier phase. GPS receivers normally figure-out timing using code phase technology, matching a binary timing signal that they generate until it matches the one from the satellite. The amount that they shift gives them the timing difference between them and the satellite, which helps them to fix their location.
However, the binary code signal operates at a relatively low frequency; the ones and zeros that make up the signal only flip over relatively slowly. That means that they can only match as closely as the frequency allows. The frequency of the radio signal carrying the code is far higher. Capable receivers can use this to improve their signal, creating more accurate timing and pinpointing their location to within a couple of millimetres.
The US system was the first GPS constellation, but it is not the only one. It is part of a broader category of systems, known as Global Navigation Satellite Systems (GNSS). Several other systems are either planned or deployed around the world.
GLONASS (Globalnaya Navigatsionnaya Sputnikovaya Sistema), developed by the Russian Federation, was born from military requirements during the later Cold War era. The system became fully operational with 24'global satellites in 1995, just a year after the US system, but is said to have then fallen into disrepair during Russia's financial crisis. It is now operational again. Russian experts claim ground station-augmented accuracy at the sub-two-metre level.
Russia stole the jump on Europe, which is meanwhile preparing its Galileo system now being built. The project, controlled by the European Space Agency (ESA), is EU funded to the tune of '5bn. It will see 30 satellites launched into orbit (27 operational and three in reserve), along with a ground control infrastructure. The angle of the satellites will provide coverage all the way up to the northern polar regions, ESA says.
Aviation-level accuracy is provided by the European Geostationary Navigation Overlay Service (EGNOS), which is the European version of WAAS (Wide Area Augmentation System – the air navigation aid developed by the Federal Aviation Administration to augment GPS), and which is already operational.
So far, four experimental satellites have been launched for orbital validation. The agency hopes to have 18 aloft by 2015, with the rest in orbit by 2020. Galileo will provide three initial services in the 2014-15 timeframe, says the EU: a free, open public service; a restricted service with an additional band for specialist users such as government bodies; and then a search-and-rescue service.
Potential users can expect full operational capacity by 2018, according to Guenter Hein, head of the Galileo operations and evolution department at the ESA: "Galileo could be more accurate than GPS because of the larger frequency bands of Galileo," Hein explains. "Galileo consists of more satellites (30) than the nominal configuration of GPS (the so-called 24+ GPS constellation). Also Galileo satellites carry the most accurate clock in space ever flown, the passive H-MASER."
Galileo also has "the opportunity to do dual-frequency measurement at no cost," points out Michele Bavaro, founder of Italian GPS consultancy OneTalent GNSS. Receivers that have access to two frequencies can use them to minimise ionospheric interference. The US GPS historically had only one civilian frequency, reserving the second (and therefore dual-band capability) for the military.
GSM systems in China
Europe may be hoping to trounce the US's older system, but it is still somewhat behind the curve: China is well ahead of the EU with its own system, called the BeiDou Navigation Satellite System (BDS), first launched in October 2000. This consists of two satellite constellations: a test constellation, originally launched in 2000, and a newer BeiDou-2 network. Like Galileo, this system will eventually put 30 satellites up in space (China has set itself a deadline of 2020). Unlike the Europeans, however, China already has 16 satellites up, and the service has also opened up to the public in Asia-Pacific, according to spokespeople.
Like other GPS developments, BeiDou is being presented as an alternative to the US GPS, which China has relied on over the years. BeiDou's developer, the China Satellite Navigation Office, wants it to command 15-20 per cent of the location-based services market in the region by 2015. The 30-station BeiDou Ground Base Enhancement System (BGBES) was approved by Chinese evaluators in March 2013. It can resolve accuracy to within 3cm using tri-band real time positioning technology.
India's GPS system – the Indian Regional Navigation Satellite System (IRNSS) – will be regional only, and it is relatively late to the race. It was supposed to be operational by 2012, but the first of its seven satellites is now scheduled for launch this year, offering 10m accuracy over land, and 20m over sea. It will also use its own GAGAN differential augmentation system for aviation purposes, which will hopefully be operational by the end of 2013.
Lastly, Japan is developing an augmentation system designed to complement the US GPS. Its Quasi-Zenith Satellite System is designed to augment existing satellite signals down to sub-metre accuracy, while also providing some message-relaying capabilities. It will complement the country's Multi-functional Transport Satellite-based Augmentation System (MSAS).
GSM compatibility issues
How compatible will the primary GNSS initiatives be? "If you have sufficient system engineering, you should be able to use any four satellites [from any system]," says Bradford Parkinson, "but there are many engineering details to be worked out."
One of the biggest problems is getting everyone to talk to each other during systems design, because once the architecture is in place and the satellites are up, it is more or less impossible to change. "Each system developed on its own, using its own scientists and engineers, and in part that's resulted in different signal structures and formats used by the multiple systems," says Richard Langley, professor in the department of geodesy and geomatics engineering at the University of New Brunswick.
GPS for mobile apps enhancement
One asset here is the International Committee on GNSS, a UN organisation that promotes information sharing. "That will be of benefit not just to the authorities, but also the manufacturers," says Langley. That is a voluntary committee, however. The International GNSS Service, provided by the International Association of Geodesy, also works closely with regional ground-station operators to combine data from all the available GNSS networks, under a project it calls the Multi-GNSS Experiment.
The next step in geopositioning technology will be to enhance the consumer-based receivers on the ground. Smartphones may currently give you enough accuracy to tell you what part of the street you are on, says Todd Humphreys, director of the University of Texas at Austin's Radionavigation Laboratory. Speaking at a TED Talk in June 2012, he said that the only reason phone manufacturers were not including carrier phase differentiation in their receivers was because of a lack of imagination. "They are not sure what the general public would do with geolocation so accurate that you could pinpoint the wrinkles in the palm of your hand," he says.
What are those applications? Augmented reality apps could position virtual objects with sub-centimetre accuracy, he argues. We can already see how that may be used in a device such as Google's Glass headset, which provides a heads-up display. Indeed, the company recently filed the patent 'Panoramic images within driving directions'.
What would it take to get centimetre-positioning inside something that wasn't a phone? Humphreys predicts a version of the 'GPS dot', as seen in 'The Da Vinci Code': a tiny receiver, less than a centimetre across, and locatable within two feet, anywhere on the surface of the globe. He anticipates sticking them to everything worth more than a few dollars. "I should be able to ask my house where my socks are," Humphreys said in his TED Talk.
GPS chipsets are reducing rapidly in size. Michele Bavaro at OneTalent GNSS says that one breakthrough came in 2005-06, when higher-sensitivity receivers came to market, making it possible to detect lower-decibel signals that made receivers easier to use in cars and opened up the auto navigation market to a general audience.
More recently, he has seen dual-receiver GLONASS/GPS chips no larger than two grains of rice. "Now you can find GPS receivers that have two separate radio-frequency chains, with very complex DSP inside to combine measurements from multiple constellations together, drawing very little power from the host," he says. They start at $15, and are far cheaper.
The sticking point for OneTalent GNSS's Michele Bavaro is antenna technology, which he says has not made the same progress as the GPS receiver: "The fact that you can now use a very little antenna is only possible because your receiver can work with very low signal-to-noise ratios."
What does all this mean for practical applications? "If you're talking about [a mechanic or builder] tracking their tools, then it's harder, but people are working on it," Bavaro explains. "One big issue is, if you go indoors, signal strength is attenuated so much that you have to use some tricks to dig that signal back out."
Signal attenuation and multi-path interference (where satellite beams bounce off buildings) make it hard to get a good GPS fix, either in the 'concrete corridors' of major cities, or indoors. Several alternatives have been proposed to harvest 'opportunistic' signals, including the use of local Wi-Fi access points, which relies largely on received signal strength indicators to determine the distance from the access point.
Traditionally, neither transmit timing or positioning data, meaning receivers have to rely on their own calculations to determine their location. Wi-Fi triangulation maxes out at around 10m, says Langley, if you're lucky. On the other hand, the signals are stronger, making them more usable indoors.
"The next positioning technology will be LTE, in my opinion," argues Bavaro. Release'9 of the 3GPP mobile communications specification gives cellular service providers the capability of transmitting timing data in the reference signal. "The cellphone will triangulate cells, and when the bandwidth is high enough, LTE can achieve accuracies similar to GPS. "It has its limits, but LTE is going to be there," Bavaro adds. "It's still to be seen if you can get one metre."
Meanwhile another company, Australian-based Locata, promises accuracy to within 1.5cm by doing away with satellites altogether and using ground-based transceivers. Receivers communicate with them, and use them for relative positioning. CEO Nunzio Gambale says that the systems, which shipped commercially in October 2012, are currently serving specialist applications such as mines, where machine control requires high-precision applications without any break in signal availability.
With all this innovative activity, what's next for conventional GNSS? The US is at present modernising its own system, launching four new signals at two new frequencies. Parkinson hopes that differential receivers will continue to enable new applications in areas such as machine control and aviation.
The US will continue spending $1bn each year on a service that is free for everyone around globe. With a world now largely dependent on GPS in some form or respect, some observers might ask, 'does it really have any choice?'