There is little evidence to support the idea that a single cellphone left on in a piece
of carry-on luggage could adversely affect instruments in the cockpit. But if dozens of phones–along with wireless-enabled laptop computers and other power-emitting portable devices–were to be intentionally switched on in flight, the resulting cacophony of electronic noise could cause dangerous levels of interference that would almost certainly send navigation sensors into tailspins under the right circumstances.
That was the conclusion of researchers at Carnegie Mellon University in a study that sought to determine once and for all the probable risks associated with using portable electronic devices on board commercial airplanes. Researchers determined that while the danger passengers’ cellphones and laptops currently pose is probably relatively small, the proliferation of such electronics could create serious safety concerns in the future.
The study’s findings are especially relevant in light of plans by airlines to install equipment allowing passengers to use their personal mobile phones in flight. Air France last month said it will soon take delivery of an Airbus A318 fitted with a system designed to permit airborne mobile-phone calling, routed through the airplane’s satcom system. The airline plans a six-month trial of the service early next year to gauge passenger interest.
The flight tests will involve the installation of an onboard “pico cell” in an overhead baggage compartment. This receiver antenna will act like a regular ground-based cellular tower. But because this cell will be located in the cabin close to callers, the phones will automatically switch to their lowest power settings, ensuring that the transmissions do not interfere with real cell towers below or affect cockpit avionics.
At least that is the theory. OnAir, a company created jointly by Airbus and SITA to develop the pico cell technology, claims that because its mobile communications system requires much lower power, the potential for interference with navigation and other sensors is all but eliminated.
The lead researcher on the team that conducted the Carnegie Mellon study said the assertion that passengers’ phones will operate at safe power settings in the presence of a pico cell is true–but only up to a certain point.
“I don’t oppose what they are saying in theory,” said Dr. Bill Strauss, the study’s author. “The good news is a pico cell is going to tell everybody’s phone to go to low power because it can hear them fine. But here’s the bad news: because you will likely have many carriers making a contract with one service provider, you now have a situation where passengers will walk on board the airplane with a particular cellphone that is not GSM. It’s CDMA or whatever. With the current technology, the pico cell will have no ability to tell that CDMA phone to be quiet. So, yes, it’s true, all of the GSM phones that are compatible with that system will go to low power, but those other phones will be screaming at the top of their lungs trying to reach a tower that’s 35,000 feet below the airplane.”
Cellphones operating at high power can interfere with ILS, VOR and GPS signals, and the Carnegie Mellon researchers have presented the data to prove it. With the assistance of the FAA and three major U.S. airlines, Strauss and his colleagues installed spectrum analyzers on 37 revenue flights along the Eastern Seaboard in 2003. The group’s findings showed that not only are passengers trying to use their cellphones with regularity in flight, signal activity was often observed in the GPS band at levels that could cause dangerous interference.
Strauss said that on each of the flights on which data was taken, at least one phone was left on, and usually several were operating. Normally the phones appeared to have been left on unintentionally, posing little danger of interference, Strauss said. But the data also showed that passengers actually tried placing calls while airborne, on average one to four times per flight.
“On every flight we detected some kind of call energy,” Strauss noted. “So people are using their phones with regularity in violation of FCC rules.” In fact, Strauss said he personally saw a young woman on one of the flights talking on her cellphone during the takeoff roll and initial climbout.
Strauss added that GPS interference was the most prevalent of all types of electronic noise observed during the flight tests, conducted aboard Boeing 737s and an Airbus A320. Normally, the researchers should not have detected a noise signature in the GPS frequency band because the satellite signals are so weak–weaker, in fact, than their equipment was capable of measuring.
But it turns out that on one in three flights, the team saw interference, and they saw it at fairly high levels, Strauss said. Further analysis of the data showed that the interference in most cases was caused by cellphone use, based on a correlation of when calls were being placed and when GPS signals were interrupted.
It’s interesting to note that under normal circumstances cellphone transmissions and the GPS frequency band do not overlap. Why the concern about cellphone use, then? All electronic devices emit radio frequencies. This stray RF normally gets absorbed by the device itself or, if it can’t be eliminated, manufacturers try to move the RF signals to non-critical bands. The trouble is, it’s possible sometimes to observe non-intentional emissions from cellphones directly in the GPS band, caused either by design flaws in a particular phone or by what is known as frequency intermodulation. Intermodulation is when frequency A from one unit and frequency B from another unit combine to cause frequencies C, D, E and F. Somebody could be using a cellphone that’s on a frequency well outside the GPS band, but that signal could combine with a frequency that’s being emitted from some other passive system–say, a laptop computer and energy is presented in the GPS frequency band.
That’s not the only way a cellphone can interfere with GPS signals. NASA’s Aviation Safety Reporting System contains dozens of reports from private pilots, and at least one business-jet flight crew, of losses of GPS signals while using cellphones in flight–illegally, it is important to point out. Responding to the reports, NASA conducted laboratory tests and found that a specific model cellphone that was cited most often in the reports emitted signals in the middle of the GPS band.
When NASA researchers investigated further, they concluded that according to the FCC standards cellphones are designed to meet, there was no violation. The manufacturer of the phone in question did everything that was required to make sure the device was safe–and yet for whatever reason, this particular model cellphone caused GPS signal interference.
After its investigation, NASA noted that if this cellphone had been a piece of avionics equipment, it would not have been permitted on the airplane. But because it was a portable device brought on board by a passenger (or pilot), it was perfectly legal– and yet a potentially serious safety risk.
What Should Be Done
There is evidence that cellphone activity can affect other onboard sensors as well. The Carnegie Mellon research team noted interference in the VOR and ILS bands, writing that onboard cellular activity is widespread during all phases of flight in spite of FCC and FAA regulations prohibiting it. The team even observed signal activity during approach and landing at levels it concluded could interfere with GPS, and on a number of approaches recorded electronic noise in the VOR/ILS bands as well.
To alleviate the potential risk of portable electronics inadvertently interfering with avionics, Strauss cited a number of actions that, in his opinion, ought to be taken immediately. First, he said, the FCC and FAA need better harmonization of standards to ensure that consumer electronics brought on board aircraft don’t present a potential danger that no one realized existed.
Next, he said that by installing some basic and inexpensive receiver technology, the pilots could be given an indication in the cockpit, say during approach, that a high level of interference existed. In turn, they could make a passenger announcement emphasizing that portable electronics must be switched off immediately.
Finally, Strauss said NASA and others should spend more money exploring the potential risks of portable electronics on airplanes. While it is fairly easy to make a piece of avionics equipment fail in the lab by holding a cellphone near it, he said, nobody yet knows the real effect such devices have on cockpit sensors.
Such tests, however, shouldn’t simply involve a series of NASA flights carrying a cadre of cellphones and laptop computers, Strauss said. That would ignore potential dangers that might arise in airplanes suffering from anomalies with their avionics equipment. For example, when an airplane undergoes maintenance, it’s possible that one of the grounding straps that controls interference might not have been reattached correctly, or perhaps a surface that was supposed to remain bare to allow for bonding contact was painted by mistake.
It comes down to the fact that a woefully inadequate amount of information currently exists to give researchers definitive data to make intelligent decisions, Strauss said. “Let’s say two years down the road we have an accident, and then in another year we have another accident,” he said. “How are we going to attribute those accidents to cellphone use? Right now as things stand, we really have nothing to do that. If we want to prevent that one accident that might occur every five or 10 years, then we need to start concentrating some energy in this area now.”