LightSquared's Proposed Fix for GPS Interference Requires New Testing

By Wayne Rash  |  Posted 2011-06-30

LightSquared's Proposed Fix for GPS Interference Requires New Testing

The long-awaited 750-page report of the GPS Interference Technical Working Group released June 30 confirmed what multiple government agencies and private industry interests have been saying since the group began its work: that LightSquared's plan for a wireless broadband network would prevent GPS devices from working properly anywhere in the United States.

But even before the official release of the Technical Working Group Report, LightSquared went public with a proposed workaround to the GPS interference problem.

LightSquared is proposing that it use the lower 10MHz of the L-band spectrum the company was given for its 4G LTE service and that it reduce the power to 10 percent of what the Federal Communications Commission originally authorized. According to LightSquared Executive Vice President Martin Harriman, this would mean that the LightSquared transmitters would only produce about 1,500 watts of radio power, versus the 15,000 watts the FCC originally authorized.

It's worth noting that the test transmitters that the TWG used in its study were running at about 1,500 watts of power, but they were using the upper 10MHz of the frequency allocation as well as the lower 10MHz section. There has been little formal testing of GPS interference using just that lower frequency slice. Because of this, many of the Working Group tests would have to be rerun to see if the interference problem has been solved.

The problem is one of basic physics and radio design. The GPS frequencies span 20MHz of the L-band immediately adjacent to the upper 10MHz segment granted to LightSquared. A strong signal in an adjacent channel can cause the receiver's RF amplifiers to become saturated, even when the strong signal isn't on the same frequency.

This can cause receiver desensitization, and it can create spurious signals inside the receiver. The result is that radio reception in the affected receiver may be impossible, degraded, distorted or some combination of the three.

Compounding the problem is that you can only do so much about it. GPS receivers have to be very sensitive to pick up the faint signals from the GPS satellites. Because they are so sensitive, they are also prone to interference. You can design in what's called a band-pass filter to block some of this interference, but a band-pass filter introduces its own problems, including reducing the signal strength of the receiver being protected. When that happens, the whole issue of high sensitivity comes to the fore again, and you're basically stuck with diminishing returns.

LightSquared Claims Fix Will Cut 99% of Interference


One solution to this problem is to move the transmitted signal farther away from the receiver being protected and another is to lower the power of the interfering transmitter. LightSquared is proposing to take both of these actions.

But will it work? Right now, nobody knows for sure. LightSquared is convinced that it can eliminate interference to 99 percent of all GPS receivers. That still leaves a lot of GPS receivers that will be affected, and most of those appear to be high-precision devices used in science, surveying and agriculture.

The U.S. GPS Industry Council argues otherwise, saying in their report that LightSquared should not be permitted to use the L-band spectrum for their data network. But their conclusion is based on the findings of the Technical Working Group, which basically said the same thing.

The Industry Council refers to problems in limited testing of the lower 10MHz portion of the L-band granted to LightSquared. The group said that 20 of 29 devices tested displayed "harmful interference." The group doesn't define what harmful interference means in this case.

Where that leaves us is that LightSquared thinks its solution will work, while the people backing the industry groups think it won't. Right now, nobody knows for sure who is right. Clearly there needs to be more testing, and LightSquared has already said it will hold off on deploying its system until it's clear that there will be no interference.

Fortunately, the FCC doesn't have to make a decision on this until testing of LightSquared's final system shows that it doesn't interfere with GPS. How difficult that will be for LightSquared remains to be seen.

But in the meantime, there remain other options that might help. For example, LightSquared's transmitters operate at 1,500 watts. Do they need to use this much power? Cell towers, by comparison, average about 100 watts each and manage to reach their customers with voice and data service.

Can LightSquared crank down the power knob and still work? Again, we don't know because no tests have been run. Can LightSquared move to a different frequency that's not next to GPS? Again, we don't know, but the L-band was originally intended for satellite use, rather than terrestrial use, and that's part of the problem.

So right now the answer is that there is no answer, and there won't be until more testing is conducted. 

Rocket Fuel