Son of Timatollah

I was on Fox 35 Orlando's morning smile show, "Good Day", for less than two minutes talking about the news that the FAA was going to take another look at allowing people to use their iPads below 10,000 feet. Well, I thought this is what I was going to talk about. The host I was interacting with only wanted to know whether she would be able to use her cell phone. She was disappointed when I told her it was unlikely she would be able to use her cell phone. (No video available at this time.)

As to PEDs and T-PEDs (Portable Electronic Devices and Transmitting-PEDs), I know it's hard to keep in mind, but excepting some of the satellite-navigation stuff like GPS and ADS-B, the designs of communication, navigation, and surveillance (CNS) electronics in certified aircraft have been in place for decades. VOR—VHF Omnidirectional Ranging—was invented in the 1940s. And none of it, even the new stuff, was designed thinking that planes would be full of radio-frequency emitting devices; you can make an argument that the more recent stuff should've been, but it wasn't. And there are about 20 or so different CNS technologies that the ones on any individual aircraft could be chosen from.

There are two real technical issues: The problem with T-PEDs isn't their operation at the frequencies they're supposed to work at: the avionics will reject those frequencies like they ought to. But RF from multiple T-PEDs can get picked up either by equipment in racks in avionics cabinets or by cables (yeah yeah, they oughta be better shielded) running the aircraft's length: that's called getting in the back door. Intermodulation between different frequencies due to nonlinearities—from bimetallic junctions to transistors—can generate frequencies that are in the bands the avionics are looking for. For legacy stuff, that can look like signal of interest, so if the intermod frequency is one of the two frequencies used to determine where the centerline of the runway is for an instrument landing (the localizer: one frequency is stronger when you're to the left; one when you're to the right; measure the strengths and compare), then you might not be on center on approach. For more modern stuff like GPS, you've got more energy there that's not the signal you're looking for, so the ability to capture and decode the GPS signals is degraded.

For PEDs—and when that Kindle is doing a next page eInk draw, it's probably emitting RF, albiet small. And it's still got to be sitting there idly running some loop at however many MHz waiting for your next-page gesture unless they've got that sucker full of discrete logic or an FPGA equivalent implementing the user interface—the emissions are sufficiently small that coupling into the avionics by the back door is almost certainly ignorable. The problem is that that RF, and that's pretty broadband RF noise btw, can get picked up by external antennas on the aircraft—i.e., go in the front door. When that happens, that's going to increase the amount of power that's noise as far as the avionics is concerned, which reduces detection or discrimination performance in analog electronics and increases bit error rates for digital stuff.

There are examples in the Aviation Safety Reporting System where there have been avionics weirdness, and then where passengers—sometimes pilots—have turned personal electronics off and the weirdness went away. Yes, correlation isn't causation, but the vast majority of commenters at tech blogs on this issue seem to pretend that there's never been a single event where a powered-up PED influenced avionics behavior. That's just not true.

The RTCA, a non-profit membership organization that works with the FAA on avionics specs, had a special committee with membership from avionics vendors, aircraft manufacturers, pilots, flight attendants, electronics manufacturers, wireless trade organizations, etc., looking at this stuff several years ago. They wrote one document, RTCA/DO-294c, "Guidance on Allowing Transmitting Portable Electronic Devices (T-PEDS) [sic] on Aircraft", 16 DEC 2008, which tells airlines how they can approve PEDs for use onboard. Everyone gripes that the expenses to do what they're suggesting are too much. They also wrote RTCA/DO-307, "Aircraft Design and Certification for Portable Electronic Device (PED) Tolerance", 11 OCT 2007, (with Change 1 published 16 DEC 2008) giving criteria for RF immunity for back-door T-PEDs and path loss for PEDs such that aircraft could be PED tolerant.

My guess is that the FAA is going to issue an Advisory Circular telling manufacturers how they can beef up their aircraft to be more RF immune and PED tolerant. If they want to change the regulations, they have to go through the rule making process. I'm really skeptical that they're going to make any big changes in any short time. 

It's a real issue, and the FAA's safety focus means they won't move fast and without real evidence that no harm is done by the change.