Capital Currents - FCC is Tilting at Windmills
... when it comes to VoIP location determination.
The term “tilting at windmills” has two different meanings. Don Quixote was trying to fight battles with imaginary enemies, for one. But the other meaning is fighting futile or unwinnable battles. That’s a good description of the current status of FCC interest in voice over Internet Protocol (VoIP) location determination. And here’s why.
The FCC is fighting a futile battle. The battleground is VoIP phone service, and victory would be automatically determining the caller’s location when an E911 emergency call is made. But it’s just not technically feasible, at least for nomadic users. The FCC in July started a new inquiry and is once again hoping for some breakthrough. I first wrote about this problem in 2003, and technical limitations have not changed much.
The first problem arose when monopolistic telephone companies claimed that VoIP was not telephone service, and refused even to connect 911 calls to public safety answering points (PSAPs). The FCC in 2005 ordered phone companies to interconnect and required VoIP providers to notify customers that their VoIP service did not have the same E911 capabilities as wireline phones. Then VoIP providers were required to have fixed VoIP customers register their locations. So today, VoIP services like Vonage and Skype support emergency calling, providing the customer registers the location in advance.
The problem arises with nomadic VoIP service. The FCC recently reported that “commenters agree that at this time, there is no technological or cost-effective means” to determine the locations of VoIP users. If you don’t register your location, or if you move from the registered location, your location cannot be determined.
Nonetheless, the public safety representatives have urged the FCC to make automatic location determination mandatory, regardless of technical feasibility. But the FCC went down that road years ago and got into trouble. In 1978, in order to promote the use of UHF television channels, the FCC required that all new TV receivers have a UHF tuner with a noise figure of 12 dB or better. But the FCC admitted that a 12 dB noise figure was not achievable at that time and was “based on our forecast of future technical developments.” The EIA’s Consumer Electronics Group (now the Consumer Electronics Association) appealed, and the US Court of Appeals overturned the FCC on the basis that the requirement “was not attainable under present technology.”
There have been a number of location determination technical proposals, but they all have problems. For example, some VoIP services appear as applications on cellular smartphones. Many smartphones today have GPS receivers built in, in order to comply with E911 location requirements that apply directly to the cellular service. But the VoIP service is an “over-the-top” service, so the cell phone operator does not know when the customer is using VoIP to make an emergency call, and the VoIP application does not have access to the GPS data in the phone. (It’s hard to imagine why anyone would use the VoIP application to make an emergency call rather than simply dialing 911 on the cell phone itself.)
Conceivably, that concept might be applicable to VoIP applications that reside on laptop computers connected to the Internet via wired Ethernet networks. The Internet service provider presumably knows the location associated with the IP address assigned to an Ethernet local area network. But, here again, the ISP would not know when a VoIP emergency call is being made. And there is no standard way for ISPs to supply location data to PSAPs.
Relatively few laptops have GPS receivers built in. But many laptops connect to the Internet using Wi-Fi, and companies like Google have been compiling private databases of Wi-Fi access point locations. For at least the last five years, Google’s Street View cars, and others like Skyhook, have been driving our streets and recording the characteristics and locations of Wi-Fi access points. So, in principle, a PSAP could interrogate a server when an emergency call comes in and determine the caller’s location. That might work on NCIS and other TV programs, but the E911 system does not have that capability today, and, anyway, the Wi-Fi location data resides in privately owned databases.
What happens when I dial 911 from my laptop VoIP service? I don’t know; I’ve never tried it.
So this whole FCC battle is futile, both for technical reasons and for jurisdictional reasons. It’s hard to imagine that it will ever be technically feasible. The FCC is tilting at windmills.