Whenever you know your position, it is in relation to one or more other objects. For example, if you travel in a straight line away from your home, your car’s trip odometer can tell you how far you are from it; by taking bearings on two landmarks with a handbearing compass, a sailor can find her position at the intersection of those two lines on a map; a Global Positioning System (GPS) receiver calculates its position in three dimensions by trilaterating its distance from four satellites (geometrically it would only need three, but the fourth one is necessary to compensate for the relatively low accuracy of the receiver’s clock).
Some positioning systems make use of natural objects or phenomena. For example, celestial navigation uses the stars, compasses use Earth’s magnetic field, and ancient mariners relied on trade winds. Other positioning methods rely on facilities or networks built for the purpose — such as lighthouses, buoys, or GPS satellites. Yet a third way is to exploit as “beacons of opportunity” devices that were built and deployed for other purposes. This is an emerging trend to compensate for the deficiencies of GPS receivers — which work very poorly in “urban canyons” and not at all indoors — as an increasing number of consumer, business, and government applications require location awareness.
Skyhook Wireless has developed a WiFi positioning system (WPS) that relies on the growing popularity, in U.S. urban areas, of wireless Internet access points. It uses them to determine the position of mobile WiFi-enabled devices, such as laptop computers, PDAs, and cell phones, on which users have installed the company’s free software. Skyhook’s fleet of GPS-equipped vehicles drive US streets and map the locations of hundreds of thousands of wireless routers. They also collect the routers’ unique ID numbers, as well as signal strength and other data. In turn, each time a device uses the system, it reports the IDs of the routers whose signal it captured. This makes the system self-correcting, because it can detect when a new router has been installed or an old one has been turned off or relocated.
Rosum Corporation, on the other hand, has developed a system that uses television broadcast towers as beacons. I discussed Rosum’s technology and business strategy with Todd Young, the company’s Vice President of Marketing. Rosum, he told me, is currently focused on two primary markets: femtocells and mobile TV location applications.
Femtocells are essentially small local versions of wireless base stations. This is a promising technology because it allows cell towers to hand off connections to femtocells, thereby freeing up frequencies and lowering operating costs. However, femtocells require very accurate in-building timing and location. Rosum’s system achieves this by trilaterating off of the timing signals broadcast by television transmitters — the location of which is precisely known. Additionally, Young points out, consumers, wireless carriers, and the FCC, will expect 911 calls dialed from phones carried by femtocells to provide location to public safety answering points (PSAPs).
Rosum works directly with its vendor partners to integrate its solution into their femtocell boards. “We provide them a reference design and some software that runs on the local processor,” says Young. This spring Rosum signed an agreement with 2Wire, a company that makes home media gateways. 2Wire has agreed to base its femtocell products on Rosum’s FemtoSynch, a hybrid TV-GPS solution that provides frequency stability, network synchronization, and E-911 location. Rosum will be starting trials with 2Wire in a few months and expects initial deployments by late 2009.
Rosum’s tests, Young says, show that its system gets about twice the ranging accuracy from digital transmissions as it does from analog signals. Therefore, the switch to digital-only television broadcasting in the United States, next February, will not hurt Rosum, despite the fact that some smaller transmitters will go off the air.
With regards to mobile TV, “the sky’s the limit,” says Young, as TV tuners are integrated into a growing number of mobile devices — such as laptops, ultra-mobile personal computers (UMPCs), and cell phone handsets. “The beauty of those platforms,” Young says, “is that everything that we need to produce location is there and our solution is a software application that runs on the device and turns that tuner into a location device that works indoors.” While “the endgame is all about handset location,” he cites many other possible applications, such as geofenced data access on laptops.
A demonstration in the UK with National Grid Wireless this summer, says Young, proved the ability of Rosum’s system to use DVB-H, which is a mobile TV standard deployed in Europe and in Asia, as a beacon of opportunity for location. While the company is initially focusing on the North American market, it will follow mobile TV standards in Europe and Asia.
Compared to other positioning systems, Young says, Rosum’s competitive advantage is in availability. “It is truly a wide-area solution. In a metro area like San Francisco, GPS coverage is essentially all the outdoor areas and about a meter inside the major buildings in the downtown area. Wi-Fi coverage is basically the areas that you can see from the road.” On the other hand, TV location coverage in the Bay Area, he claims, is available and accurate “on any random floor in any random building,” including the subway.
Additionally, Young points out, the broadcast TV infrastructure that Rosum uses for location is the backbone of the emergency alert system — with back-up power and great redundancy. “If you’re going to do 911 safety-of-life applications or security applications, where it is critical to know where someone really is, then GPS is out of the game because it does not provide indoor coverage. WiFi is out of the game because you cannot have a security system that goes down when the power goes out and their database depends on people inputting accurate data. So, in terms of availability, coverage, and robustness, we think that we clearly have the best solution available.”
In an article titled “Femtocells: Bringing Reliable Location and Timing Indoors” in the 2007 September/October issue of Inside GNSS, Young and Dimitri Rubin lay out the technical details of Rosum’s system and summarize the advantages of television broadcast towers as beacons of opportunity:
- TV signals are broadcast across every metropolitan area on Earth and can typically be used at a range of up to 50 to 100 kilometers from the transmitter
- The low frequencies and wide bandwidth used for TV broadcasts allows the signals to penetrate buildings well and mitigate multipath effects
- TV signals have stable timing and are highly reliable; in the United States, broadcasters’ participation in the Emergency Alert System requires them to maintain service even during disasters
- The new signals from mobile TV networks can supplement those from existing broadcast TV networks; TV location and timing information can also be combined with other signals, such as GPS, to create hybrid solutions that perform better than any single solution
- Where users have clear line-of-site visibility to regional TV towers, TV-based positioning is as accurate as GPS
- TV and GPS are very complementary positioning systems, because TV transmitter geometry is excellent in dense urban areas and indoors, while GPS works best in unobstructed rural environments, where there are fewer TV transmitters.
By: Matteo Luccio, President, Pale Blue Dot Research, Writing, and Editing, LLC
what’s also interesting is considering the possibility of positioning *without* interpreting the content stream that’s being received. that is, if a receiver is in an environment with certain static sources in range, it should be possible to calculate 3-space bearings to each. it may be possible to monitor signal strength to derive a pseudo-range, although i’m less optimistic about that.
anyway, the point is that these bearings could be converted into a set of barycentric coordinates. if there are enough static sources, least squares could be used to find position.
the problems of secondary, non-static sources, like mobile transmitters, might be solved using either “semi-interpretive techniques” or by adapting multipath filters like are used at airports for positioning by ground radar. the “semi-interpretive techniques” might parse enough of a packet stream to identify the source of something on a channel, and then discard the rest.
3-space bearings might be determined by using inteferometers which measure conic angles mounted on the receiver.
if the receiver also has some other independent means of determining absolute position, it might be able to self-survey these passive sources and estimate their absolute positions.