Locating became a standard feature of mobile phone service in the years following the Federal Communications Commission’s (FCC) order released in 1999 that described in detail how mobile network operators must provide the locations of wireless 911 callers to emergency call centers. The decision to require wireless 911 locating was first announced in 1996, when mobile phones were mainly used outdoors. People still relied on wired phones indoors — and the phone company already knew the street address associated with each pair of wires.
Flash forward 16 years. A growing percentage of people use mobile phones exclusively. Most 911 calls are made from wireless phones, and most wireless calls are made indoors. That’s a problem, because the two solutions embraced by mobile operators to meet the FCC’s original wireless 911 mandate (one handset-based, the other network-based) are great for locating callers outdoors, but terrible for locating callers indoors. Even worse, indoor locating demands greater accuracy than outdoor locating. Emergency responders can usually spot an accident on the street once they get within several hundred feet. Indoors, even a small error could put emergency responders on the wrong side of a wall.
The problem became clear when data obtained under the Freedom of Information Act showed that over a 90-day period in Washington, D.C. only about 10% of more than 90,000 wireless 911 calls provided accurate location information. Stories spread quickly about 911 callers dying. In one case, a kidnap victim secretly dialed 911, but police dispatchers could only listen helplessly for 18 minutes as he was repeatedly stabbed.
What the FCC should have learned from this experience is that the evolution of technology and tech markets is highly unpredictable. Mapping out plans six or eight years into the future is risky. Most of what was achieved with the FCC mandate probably would have been achieved without it. The wireless industry was already convinced that there were vehicle locating business opportunities.
When the FCC first considered issuing a wireless 911 locating mandate, it assumed that the solutions would be network-based. However, companies such as SnapTrack were busy developing handset-based solutions — integrating GPS receivers and mobile handsets — and the FCC had to be politely reminded not to craft rules that favored some technologies over others.
Early this year, the FCC published rules requiring increasingly effective indoor locating (within 50 meters). Vertical location data must be provided within three years. The long-term goal is to provide “dispatchable” location information such as street address, floor number, and even apartment or suite number. A series of deadlines have been defined for the four nationwide mobile operators. Smaller operators may apply for extensions.
Despite good intentions, the FCC has once again set itself up for failure. The FCC continues to hold cellular operators responsible, as if nothing has changed since 1998. Most of the progress in indoor locating now comes from other companies. Cellular operators have less control over handsets since the introduction of iPhones and Android phones. And today it is possible to make wireless phone calls using Wi-Fi — completely bypassing local cellular operators.
The most accurate and reliable indoor locating systems are also based on Wi-Fi. Skyhook Wireless and Google pioneered the technology by “war driving” the streets of major cities around the world, identifying and measuring the signal strength of every Wi-Fi access point heard, and stamping each record with the vehicle’s exact location. Apple has since created a Wi-Fi database that is continuously updated using crowdsourced data from millions of handsets. TruePosition, a mobile locating company that acquired Skyhook Wireless in 2014, says tests show that Wi-Fi positioning can already locate mobile phone users to within 50 meters indoors about 80% of the time, meeting the FCC’s horizontal accuracy goal.
There are other applications that are helping to drive the development of more accurate and reliable indoor positioning solutions. However, most of these solutions are intended for private use. Venues with heavy foot traffic, such as airports and convention centers, are deploying sensor networks that detect the Wi-Fi and Bluetooth signals periodically transmitted by handsets. The information gathered can be used to plan the locations of concessions and to monitor the time required to get through security.
Brick-and-mortar store merchants are tracking shoppers to determine the effectiveness of advertising and to provide contextual engagement. There are several indoor “micro-locating” solutions that can serve this need.
One approach is dead-reckoning. Since the exact locations of building entrances are known, and GPS can identify which entrance was used, it’s possible to track users indoors using the inertial sensors found in many handsets. Rather than trying to pinpoint the user’s location on demand, it’s possible to calculate the user’s current position based on the direction and speed of travel starting at the entrance.
Another approach uses variations in the Earth’s magnetic field or Wi-Fi signals on a specific channel (Wi-Fi fingerprinting) to map the inside of a building. Consequently, this method is only useful inside buildings that have already been mapped — usually by someone determined to keep the information private. Beacons can also be used to locate handsets, but not nearly enough have been deployed to make a difference.
Everyone wants wireless 911 calls to include location information. And no one faults the FCC for trying to help make it happen. However, the FCC mistakenly believes it can order the wireless industry to produce innovations according to its schedule. The FCC also fails to realize that Internet companies, hardware manufacturers and software developers now have much more to say about indoor locating than mobile network operators.
Perhaps the FCC would get better results if it offered incentives and assistance to the broader industry rather than slapping mandates and the threat of penalties on four operators.
This commentary by Ira Brodsky first appeared at Computerworld. Brodsky is a Senior Analyst with Datacomm Research and is the author of five books about technology. Brodsky focuses on mobile solutions for payments, retail automation, and health care.