The GPS crisis: how serious is it?

By Charlotte AdamsThe Federal Communications Commission (FCC) in January 2010 gave LightSquared, a communications company in Reston, Va., that has been assigned frequencies in the mobile satellite services (MSS) band, a conditional approval to operate as a wholesaler of terrestrial-only, high-speed wireless services. Although the controversial waiver which allowed this to happen was the culmination of almost a decade of advocacy within the FCC, the final step appears to have been rushed through on the graveyard shift, according to the global positioning system (GPS) view. The approval also came through the day after the President's State of the Union message spotlighted high-speed wireless access as a national priority.Stung by surprise, the $110-billion GPS establishment rallied in opposition. The number-two officials at the Defense and Transportation departments and two U.S. senators wrote stern letters to FCC. Congressional hearings were held. Just in time GPS heavy hitters grasped the potential danger and have initiated a wider range of testing and analyses than originally envisioned. New U.S. Air Force Space Command open sky and anechoic chamber tests and analyses will be fed into the FCC process. LightSquared wants to provide fourth generation (4G) wireless services, using the 1525-1559-megahertz MSS band for their satellite downlink. The trouble is the company is authorized to use its downlink frequencies to broadcast high-power terrestrial signals from ultimately as many as 40,000 transmitters. In the past the MSS band was reserved for space-to-Earth transmissions, so that power levels in that area were low. The GPS L1 signal used by civil aviation is not far away, centered at 1575.42 megahertz.The company, working under an FCC schedule, plans to launch its network at the end of this year, pending FCC satisfaction with a June 15 report by a government/industry technical working group (TWG) on interference issues. Whether comprehensive testing and analysis can be completed by then remains to be seen.Although LightSquared has given the U.S. GPS Industry Council (USGIC), the co-chair of the TWG, verbal assurance that the company will not exceed 1.5 kilowatts (32 dBW) in the 1545-1555-MHz band, LightSquared is authorized to transmit at as much as 15.5 kilowatts (42 dBW) in its allotted frequency bands. The TWG is looking at GPS overload and desensitization issues for both levels of LightSquared transmit power.Experts say civil aviation receivers may have a hard time hearing the billion-times-weaker GPS received signal in the neighboring, protected 1559 -1610- megahertz, aeronautical radio navigation service (ARNS) band if the company's 1.5-kilowatt, high-speed wireless transmitters are turned on as currently designed. Because many LightSquared transmitters would be clustered in urban areas, interference effects could be multiplied. Aircraft taking off and landing from airports in urban areas could face impacts in the most critical phases of flight.Another reason GPS probably would have a problem is that some receivers are designed to receive the maximum GPS signal possible. They use wider filters and, as a consequence, pull in some signal from outside the GPS band. "We recognize the value of broadband service," says Bobby Sturgell, senior vice president of Washington operations for GPS equipment manufacturer, Rockwell Collins in Cedar Rapids, Iowa. "What’s not been considered here is the value of GPS." GPS is deeply embedded in the aviation, electrical power, financial, and cell phone industries. There are more than 1 billion GPS receivers worldwide. And there are more than 2,500 GPS-based Wide Area Augmentation System (WAAS) approaches in the U.S. Further, the Federal Aviation Administration's NextGen air traffic control (ATC) concept depends on GPS. Noisy neighbor?GPS operates in a quiet neighborhood, Sturgell says. By approving a noisy neighbor to move into the adjacent band, FCC made it a "bad neighborhood." The independent federal agency "effectively reallocated spectrum" without first going through a formal rulemaking process and comprehensive upfront testing, he says. "When you change the signal environment, you throw out all the rules."LightSquared proponents see it just the opposite. The company says it has been planning to move into the neighborhood for years. Filings of LightSquared (and predecessor companies) and GPS proponents date back to 2002, the company says. From past FCC actions the GPS community should have known that LightSquared was coming with high power and many base stations, it says. FCC removed the limit on base stations in 2005, established a base station power limit of 1.6 kilowatts in 2005 and increased the power limit to 15 kilowatts in 2010, says Jeff Carlisle, the company's executive vice president for regulatory affairs. He wonders what all the uproar over the waiver is about, since it does nothing to change potential transmitter issues. Further, Carlisle contends that the GPS community never raised the receiver overload issue during the first nine years of FCC activity.Huge problemThe potential problem for aviation is huge. GPS aviation receivers have been tested and certified for the kind of quiet environment that existed before these issues arose.There are several ways to address the problem. LightSquared could be moved to frequencies further away from GPS, the company could broadcast at lower power levels or filters could be applied to GPS as well as LightSquared. If GPS users end up having to modify the installed base, however, it will be a costly and lengthy process.The FCC's authorization document implies that GPS should share the burden of any interference emanating from LightSquared's authorized frequencies. Last month FCC went further, in a related matter, stating that, "in the case of GPS, we note that extensive terrestrial operations have been anticipated in the L-band for at least 8 years."If the aviation users of GPS have to modify their receivers because of interference, it could be a five- to six-year process, if they can succeed, Sturgell says. That includes two years to design, develop, and test new products, two years for the technical standard order (TSO) process and two years for certification.Further, as Sturgell points out, a new filter would change the housing and size of the system, making it difficult to fit back into the aircraft. "The unintended, unforeseen consequences are huge." Engineer's perspective"Depending on the kind of modulation that the GPS signal is using, it extends close to 1559 MHz," the edge of the authorized LightSquared spectrum, says Dr. Sai Kalyanaraman, a Rockwell Collins senior systems engineer. Kalyanaraman is also a member of the GPS industry-LightSquared TWG and the RTCA working group that FAA has asked to help analyze test data.LightSquared has developed special filters to limit the company's power output into the GPS band. The company has been required to meet -100dBW per MHz in the GPS band. That is very little power and is not expected to be a show stopper if LightSquared can achieve it. Rockwell Collins has conducted its own tests to evaluate the impact of the LightSquared transmitters. The company is confident that if the broadband network is turned on the way it is currently designed, there would definitely be GPS issues. The transmissions will desensitize and overload GPS receivers, Kalyanaraman says. The question is the extent of the damage.A GPS received signal at 45 dB-Hz -- a metric used to calculate signal strength relative to noise -- is considered strong. However, once the LightSquared transmitter is turned on, that number will decrease to 40, 35 and 32 dB-Hz, and then further downward until it loses lock on the signal. The downward progression is a function of the GPS receivers' distance from the LightSquared transmitters. At 32 dB-Hz, the performance of the GPS signal would be less than it would be at 45 dB-Hz. The TWG tests and analysis of aviation receiver effects have to assume worst-case conditions because these receivers are certified to meet those conditions. The Air Force does not guarantee a strong GPS signal all the time. Receivers designed for safety-critical applications have to continue to function under minimum guaranteed conditions -- not 45 dB-Hz, but 32-33 dB-Hz at the input of the receiver. The tests also assume the minimum guaranteed GPS constellation, which is 21 plus three orbiting spares, not the 30 satellites now in orbit. Analysts also must consider the different satellite geometries available at different times and places. TWG aviation receiver tests will be performed at the minimum, guaranteed signal level specified in the GPS signal standards document. In the anticipated noise environment this translates to about 32-33 dB-Hz carrier-to-noise ratio (CNR) at the input of the receiver, Kalyanaraman says. At this point the LightSquared signal is added to the setup.When the LightSquared signal is added to this weak GPS signal, the CNR would decrease rapidly from the 33 dB-Hz level, he says. Down at CNRs of 30, 28, or 27 dB-Hz, ranging accuracy is impacted and position accuracy would suffer. The receivers could lose lock on the signal. In addition, the receiver's ability to decode WAAS messages is severely degraded at these low CNRs.Worse still, if required ranging accuracy is degraded beyond a predetermined threshold, the integrity analyses performed for certified GPS aviation receivers may no longer be valid. "Integrity is key for certified aviation receivers," Kalyanaraman says. For a civil aviation receiver, integrity means that a reported GPS position solution is guaranteed to be within a particular radius of the truth, with a 10-7 likelihood that the position error will be greater than this radius.The TWG's aviation subgroup is conducting lab tests -- not open sky tests -- on 10 GPS receivers, using a signal generator. This includes aircraft receivers as well as WAAS reference receivers, Kalyanaraman says. The tests will measure the impact of LightSquared transmissions on key GPS performance metrics, such as ranging accuracy and WAAS message failure rate. The tests will determine at what LightSquared power level a GPS receiver has problems and then run the analytics and the math to determine the distance from the transmitter/s at which this problem will occur.