How lack of wireless spectrum delayed implementation of Amtrak’s PTC system
It’s still not clear what caused the crash of an Amtrak train in Philadelphia on May 13th. But it’s possible that a Positive Train Control (PTC) system could have prevented the crash—a system which was months away from being implemented when the train, heading into a curve at more than twice the posted speed limit, derailed, killing 8 passengers and injuring more than 200.
The main obstacle to implementing PTC, according to a December 2012 report by Amtrak’s Office of the Inspector General, has been acquisition of dedicated wireless spectrum in the 220 Mhz range needed to operate the system. PTC uses a wireless connection to inform a train of its location, local speed limits, and nearby trains, and can issue emergency braking or steering commands if conditions demand.
But according to critics, a more modern approach to spectrum usage might have made that dedicated spectrum unnecessary, speeding the implementation of PTC and ultimately saving lives.
“In the last ten years, there has been a significant revolution in the way people think about how you structure wireless networks,” says Harold Feld, senior vice president of the telecommunications consultancy Public Knowledge. “The buzz term for this is hetnets, or heterogenous networks—networks that don’t rely on just one set of frequencies and towers, but multiple different frequencies and handoffs.”
These heterogenous networks are also commonly referred to as multi-band or flexible-use networks. They’re a new alternative to longstanding dedicated frequency approaches, in which an operator such as a TV station or emergency service would send its signal out on one, and only one, stretch of the airwaves, to which it had exclusive rights.
According to Feld, the push to multi-band spectrum usage was ushered in by the arrival of the smartphone nearly a decade ago. All those iPhones created a bandwidth crunch for wireless carriers, who responded not just by snapping up more spectrum, but by allowing handsets to jump from crowded frequencies to underused ones as needed.
This is possible because even allocated spectrum is, broadly speaking, vastly underutilized. A National Science Foundation-funded study conducted by the Shared Spectrum Corporation between 2004 and 2007 found vast amounts of unused spectrum at various locales. In one dramatic example, only 13% of spectrum was utilized during the period of the 2004 Republican National Convention in New York City. In less densely suburban and rural areas, utilization was frequently under 1%.
Though that data predates broad adoption of both smartphones and WiFi, it shows the general potential for dynamic switching into unused spectrum to vastly increase overall network capacity.
Those continuing to favor single-band network solutions often argue that their approach enhances security and reliability by keeping spectrum use exclusive.
But according to Sascha Meinrath, director of the technology policy consultancy X-Lab, multi-band networks are also more secure and reliable than single-band solutions. In 2013, Meinrath was solicited for comment on PTC spectrum policy by the National Transportation Safety Board, and advocated a multi-band solution for PTC that would use a mix of GPS signals, 3G/4G phone networks, and unlicensed spectrum.
“Having multi-band solutions has been demonstrated again and again and again to be more reliable,” says Meinrath. “I feel like this is the spectrum equivalent of the global warming debate—you’ve got people who are in a mindset that you just can’t provide enough evidence.”
The FCC, for its part, states that it has no recommendation of dedicated over multi-band solutions for emergency and mission-critical wireless applications. FCC sources speaking on background said that Amtrak had not consulted with the regulator as it conceptualized its PTC network, but instead only asked for help securing bandwidth at 220 Mhz after committing to the single-band approach.
To an extent, Amtrak’s choice to develop their PTC system at 220 Mhz put them in line with an emerging industry standard. As early as 2007, a coalition of freight rail operators, operating as PTC-220 LLC, began building a shared national pool of PTC spectrum at 220 Mhz, and developing hardware tailored to that frequency. At that time, multi-band solutions were still novel.
But when Amtrak’s efforts to acquire compatible spectrum stalled out in 2011, they did not entertain multi-band options that, by then, had become widespread. Amtrak instead lobbied Congress to grant it dedicated spectrum, but those efforts failed. Once Amtrak finally located available 220 Mhz spectrum in late 2014, the FCC approved the purchase swiftly. Amtrak has stated that, at the time of the Philadelphia crash, it was only months away from activating PTC in the northeast.
“If the accident in Philly hadn’t happened,” says Feld, “The three years they wasted chasing 220 Mhz wouldn’t have mattered. The real tragedy is they ended up running out of time, and the delay mattered.”