Getting to Z: Indoor Positioning with GPS

By Alan Cameron

In this column, I normally write about satellites, signals, and space (as in outer), and the policies or controversies pertaining to those entities. This week we are headed indoors. Inner space, where GNSS has difficulties going, but must go, somehow, to prove itself commercially and governmentally. To do so, it needs powerful friends.

The most rigorous indoor location testing to date got underway two weeks ago in the San Francisco Bay Area, in trials organized by a Federal Communication Commission’s (FCC) advisory committee, the Communications Security, Reliability and Interoperability Council (CSRIC). The tests seek to lay the groundwork for future FCC rulings on indoor location requirements, to which wireless carriers must adhere. The trials run through December 31, in dense urban, urban, suburban, and rural test blocks around the Bay.

For the sake of the GPS/GNSS industry and community, whatever technology solution emerges from these trials as the favorite, GPS/GNSS had better prove itself as a part of it, not only to gain a foothold in indoor markets and applications, but to preserve its standing in outdoor environments. Other positioning technologies have sprouted up like mushrooms, filling in vacant micro-niches. The indoor environment as a whole is just that, an environment, not a niche, and where it goes — taking the money with it — outdoor may likely follow. Wi-Fi, for example, is gaining installment base by leaps and bounds, and probably currently supplies the best unaided indoor location — where it is installed.

“Retailers are desperate for more customer data, this [indoor location data] is golden,” says Janice Partyka, GPS World’s contributing editor for wireless. “They probably won’t wait for the requirements or for the wireless carriers to push out the solution. Some venues like airports can track you now. This time around, commercial uses will precede E911.”

Although the need for accuracy is arguably greater indoors, so too are the difficulties — and the costs. At stake is getting room-level and floor-level location accuracy from a mobile 911 call to emergency responders during the Golden Hour, a term used in heart-attack, stroke, and trauma situations, but which applies equally to fires, violent crimes, and virtually by definition to any sort of emergency. Responders need to know “which side of the wall” he/she/it is on, and which floor — even before they enter the building.

In the floor-level or vertical component of the location coordinates resides one of the key challenges.  The vertical or Z-coordinate in a GPS/GNSS solution has always had the lowest degree of accuracy. To be sure, the barriers imposed by steel, glass, and concrete, as well as the confusion generated by multipath in dense environments, apply just as much to the X- and Y-axes, but getting to Z (since getting from floor to floor in case a mistake is made would be most time-consuming) may constitute the largest challenge.

The FCC hosted a workshop in Washington D.C. on October 24 in preparation for the tests. The workshop introduced public-safety officials’ expectations for indoor coverage, test mechanics, the technologies under test, and more. CSRIC will draft a report for the FCC based on the test results by March 2013.

The Candidates, Please. Four companies are actively participating in the CSRIC tests, submitting their diverse indoor solutions for rigorous and repeatable performance proof: Boeing, NextNav, Polaris Wireless, and Qualcomm.

The CSRIC test bed discussions started in 2010 with seven potential technologies for Stage 1:

• Polaris Wireless (RF fingerprinting)
• Qualcomm (assisted-GPS/AFLT/cell ID)
• NextNav (Wide-Area Positioning System (WAPS) of GPS-like terrestrial beacons, described here.)
• Boeing (low-Earth orbit Iridium satellites; because much closer to Earth than GPS, hence 30-dB penetration margin; a range of Iridium solutions, some of them in combination with GPS
• CSR (AGPS/WiFi/MEMS)
• TruePosition (UTDOA)
• CommScope (DAS proximity).

The latter three have since dropped out of the testing for reasons not stated.

Polaris Wireless is the only cellular-network-based location technology provider in the tests, as all other network-based location technology providers withdrew from participation in the CSRIC trials. The trial includes Polaris Wireless’ Wireless Location Signatures (WLS), a software-based radio-frequency (RF) pattern-matching approach that requires no changes to the wireless device or the wireless service provider’s base stations. The June issue of GPS World carried an article on this technology; see “Location by Database.”

Norman Shaw, Polaris Wireless executive director of government affairs and business development, serves as co-chairman of CSRIC’s efforts on improving indoor location technology. “RF does funny things. But there are cultural issues as well. It’s natural for us to expect technology to get us all the way to the goal line. However, we often overlook the challenges. Can we deliver Z-location? And can we do it in an actionable way for the emergency responder? That person needs to know, not that the emergency is 185 meters above the ground, but the number of the floor. For this and for other reasons, you need to marry different technologies.”

“This test is a great start,” Shaw concludes. “But this test bed will need to be maintained to continue testing and to test future technologies. Additionally, a second test bed will be needed in a denser, older city, probably East Coast; perhaps Chicago or New York. We should all be aware that once the testing concludes and the regulations appear, this is the emergency service we’re going to be living with for the next 20 years.”

Ganesh Pattabiraman, co-founder, president, and chief operating officer at NextNav, adds that in addition to providing data to drive regulation, the testing “brings awareness to the public safety operators and the FCC that here are reliable technologies that can address the problem of indoor location. As opposed to 10 years ago, or even six years ago. Not just ours, but others too.”

According to the NextNav website, “For devices equipped with NextNav’s technology, when a subscriber calls 911, the first responder won’t be left guessing about where they are.  Providing a unique height capability, with vertical precision of up to 1 – 2 meters, first responders can move rapidly to the correct floor to ensure that not a second is wasted in the emergency response process. NextNav’s transmission is encrypted, secure and is available for carriers as a standalone service for E911 only. A carrier can implement the NextNav solution to enhance location performance of the E911 system separate from any decision to use NextNav capabilities as part of their commercial location-based services.”

Pattabiraman continues, “The need for accurate indoor location is greater [than for outdoor], but is the technology and the cost to the wireless carriers of implementing it up to the task?  It all comes down to economics. If we or anyone can provide a solution that is incremental, reasonably priced, and commercially viable, then we can move forward.”

Particularly, he adds, “If we can build on the existing blocks of GPS at minor incremental cost, then we see the possibility of delivering the best possible accuracy for the lowest price.”

Test Administrator and Parameters. TechnoCom, a location-technology-neutral business, is conducting the Bay Area tests. TechnoCom is an active contributor to the Alliance for Telecommunications Solutions (ATIS) Emergency Services Interconnection Forum (ESIF). The ATIS conducts long-term research that serves as a basis for CSRIC findings and recommendations. The two organizations have many of the same members, although CSRIC consists of FCC-nominated members who serve one-year terms and thus doesn’t have “the consistency needed to do good science,” in one participant’s words.

The TechnoCom test parameters consist, broadly, of: a variety of locations (environments) and building types (also known as morphology), multiple test spots in each building, and each test spot to have at least 100 test calls. Researchers are looking for an indoor ground truth accuracy of 3 meters, something that would warm the hearts of public safety responders, but a level which, other experts say privately, is highly unlikely to be implemented as a requirement.

Public safety advocates would ideally want 5 meters, to the extent of “knowing which side of a wall a heart-attack victim is lying on.” Technology vendors such as those supplying solutions for test would probably settle for a 50-meter requirement, even if their solutions can do better. That’s at least in part because they are caught between the public safety folks on the one side and the wireless carriers — to whom they must sell — on the other. The wireless carriers are the most conservative of all, and may not want anything more stringent that the current outdoor requirements: 50-meter accuracy 67 percent of the time, and 150 meters 90 percent.

TechnoCom will test the following locations:

• Dense urban: a four-block area north of Market Street in San Francisco’s financial district; as one participant pointed out, this is still not the densest urban environment to be found in the United States. For that, you have to look at older, Eastern cities such as New York or Chicago.
• Urban: San Francisco and downtown San Jose
• Suburban: Santa Clara County (malls, homes, condos and some high-rises)
• Rural: Between Gilroy and Hollister, California.

All kinds of structures, about 20, typically found in the four basic environments, will serve as test spots: high-rise, mid-rise, mall, apartment building, house, warehouse, and barn. Various test points will be sited in each as appropriate, probably at 5-floor intervals in multi-storey buildings.

Indoor Positioning Webinar

GPS World will host a December 13 webinar on the subject of Indoor Navigation. Participation is free. The time is 10 a.m. Pacific / 1 p.m. Eastern / 6 p.m. Greenwich (UK) time. Registration is free.

This is the next frontier for personal and machine navigation — and many are out there now, working diligently on it.  In just one example, a new chip fuses input from several sensors, using the best combination at any given time to maximize coverage and accuracy while keeping power draw to a minimum. This produces continuous position availability in indoor environments, as demonstrated by performance measurements in real-world test environments.

The senior product manager responsible for this development joins us to talk about the inner workings and the outer manifestations of this new solution. He’ll be joined by other guest experts to be announced.

J. Blake Bullock was senior product manager responsible for CSR’s next generation of GNSS solutions. He has now transferred to Samsung System LSI Business and is responsible for GNSS and indoor positioning solutions. He holds a M.Sc. degree in geomatics engineering from the University of Calgary, an MBA from Arizona State University, and several patents in LBS and navigation.