Rivada Networks has working Android and Apple prototypes of its enhanced location-based service (E-LBS) that is designed to improve location-accuracy information for a variety of applications, from commercial uses to 911 calls to firefighter location, according to Rivada Networks CTO Clint Smith.

In the spring, Rivada Networks received a patent for the peer-to-peer technology, which is software that leverages a variety of location techniques and artificial-intelligence algorithms to get more accurate location information, even in cases where traditional location technologies are not directly available to the mobile device, Smith said. Rivada Networks personnel recently tested the solution at a building in Brooklyn, he said.

“We’re coming in pretty consistently under a meter [for XY location],” Smith said during an interview with IWCE’s Urgent Communications.

E-LBS also provides Z-axis vertical information that is not yet as precise as the XY location information, but the readings during the Brooklyn testing were within the eight-foot threshold—the height of a ceiling in many multistory buildings—that the FCC and public-safety officials have desired from location technologies, Smith said.

Although the prototypes are working and have tested well, Smith said he is confident that the E-LBS technology can be improved and optimized. Although E-LBS is software, Rivada Networks does not want to sell the solution as an application; instead, it would be better suited in the library of mobile operating systems like Android and iOS, he said.

“Our key thing is that we know we can improve it, if we can get a better handle on some of the communication protocol stack between devices,” Smith said. “We’re really at that point where we’ve got working items that you can see. There’s no question that we’ve got a massive improvement. Once we get these other things into the operating system, we’re going to blow this thing out of the water.”

 At the heart of E-LBS is a patented, self-correcting Kalman filter algorithm, which takes myriad location-information inputs—including inputs from other devices in the area—and uses them to calculate the device’s location, Smith said.

“To me, it’s pretty sharp overall, because we’ve got three things happening with it,” Smith said. “We’ve got the fused-location piece, which is what you normally have coming from your handset—a combination of GPS, cell ID and Wi-Fi ID—and we put that into the Kalman filter. Then, we have a dead-reckoning function, and that helps. And the third piece is the trilateration, where we’re getting all of the other mobiles to communicate with us.

“We take the three values, put them into a Kalman filter and spit it out at the end. When it’s done, there’s something called a covariance matrix—effectively an artificial-intelligence node that is a weighting factor—and it’s fed back into the system, which recalculates everything all over again, when something changes.”

Smith said that Rivada Networks engineers took steps to reduce the amount of processing necessary for E-LBS, in an effort to save battery life. However, Smith acknowledged that more improvements can be made in this area.

“There’s optimization that will have to be done to take it to the next level,” he said.

A volunteer firefighter, Smith developed the E-LBS idea initially to help address issues of locating firefighters at an incident scene. However, getting the E-LBS technology in commercial mobile devices—something that can serve as a complement to enhance GPS location information and 911 location information—would pave the way to the firefighter application, he said.

“Where it really has to get is into the operating system,” Smith said. “Then the gates open to getting it on the fireground and everything else.”