At Austin airport, Wi-Fi predicts how long the security line will be
- 23 October, 2014 11:43
The Internet can ease travel concerns in many ways, including flight-delay information, maps of road congestion, and ride-sharing apps. But a Wi-Fi network at the Austin, Texas, airport can now answer one of the great unknowns: How long will I have to wait in line at security?
That information is available thanks to fairly simple technology implemented on a Cisco Systems network run by global Wi-Fi provider Boingo Wireless. It's an early example of how the so-called Internet of Things can make some parts of life easier.
Austin-Bergstrom International Airport got the nation's first airport Wi-Fi network in 2000, according to Boingo, which has run the airport's Wi-Fi since 2008. Now it's become one of the first airports to implement Passpoint, the standard that lets users of some devices get on networks and roam between them without entering a username and password. The Cisco network that supports Passpoint can also use location technologies for additional services.
Travelers don't even need to get on the network to take advantage of the security-wait warning system. A forecast for how long each line will take appears on screens right outside the security checkpoint. And any traveler who goes through security with a device that has Wi-Fi or Bluetooth turned on also helps to make the system work, according to Boingo CTO Derek Peterson. Boingo has launched the wait-sensing technology at three airports, all in trial mode, and Austin's is the first facility where it's displaying the information.
Here's how the system works: Wi-Fi devices with standard settings turned on constantly send out signals looking for nearby Wi-Fi devices and access points. Access points near the security checkpoints detect those signals and the unique MAC (media access control) addresses associated with them. Using that data, the system determines when that device entered the area of the queue and when it reached the other end of the checkpoint, after the owner finished with security.
In some areas, the airport does the same thing with beacons that detect Bluetooth signals emitted by users' devices. The unique Bluetooth ID identifies each device, so it works the same way as a MAC address. In some areas, the system uses both Wi-Fi and Bluetooth.
Despite using a unique identifier for each device, the system doesn't identify the person carrying that device, Peterson said: The airport isn't concerned with who's made it through security, just how long it took them to get through.
There's one wait-time display for each line to go through security, so before they get in line, travelers can join the one with the shortest wait. In effect, they do their own load balancing, which can minimize the wait time for everyone.
The signs help set travelers' expectations, which has reduced complaints about the wait, Peterson said. But the displays are just the beginning. All that data is stored -- again, without any names associated with it -- and can be analyzed to estimate how long the security line will be at any given time and day. Boingo has shared the information with the airport and the Transportation Security Administration (TSA), which then made changes that reduced wait times, he said. It may also share the data with airlines, which could use it to estimate wait times for their customers. Delivered well before their flights, those estimates could help them decide when to leave for the airport.
Data from Wi-Fi and Bluetooth devices makes for pretty accurate estimates of wait times, according to Peterson. With about two months of historical data, the system can build a predictive model for normal days that's about 99 percent accurate, he said. In Austin, where the system has more than a year of data built up, it can now account for holiday travel surges. But other events, like the shooting of a TSA agent that shut down part of Los Angeles International Airport last year, can't really be built into a predictive model.
Boingo envisions using the network's location capabilities for other services, too. For example, the airport could predict busy times for restrooms with a detection system similar to the one at the security checkpoints, and it could track movable airport assets such as wheelchairs by equipping them with Wi-Fi radios, Peterson said.
Additional services, for travelers who opt in to them, could include notifications and retail promotions that are based on where someone is in the airport and how much time they have to get to the gate and board, he said.
Wireless location technology has improved dramatically just in the past six to 18 months, according to Peterson."It's come a long way in the accuracy that you're able to get off of both Bluetooth and Wi-Fi," he said.
As for luggage, seemingly the most misplaced thing in airports, there may be a Wi-Fi solution for that problem, too.
"That is one of the applications that we've toyed with," Peterson said.