Charles W. Rhodes /
07.05.2011 02:00 AM
Can Terrestrial Broadcasting and GPS Co-exist in Adjoining Spectrum?

The FCC has granted use of space-to-earth spectrum to a firm that intends to construct 40,000 base stations to carry broadband to the public. This sounds like a great idea, taking some of the spectrum demands from your shoulders as broadcasters.

But this re-purposing of the 1525-1559 MHz spectrum would come at an unforeseen price. This spectrum is adjacent to the international GPS satellite navigational band, 1559 – 1610 MHz. Each of these base stations would radiate 1584 watts, maybe more. These signals could jam reception of GPS signals. GPS provides time, frequency, and location data vital to broadcasters’ operations now and even more so in the future—for example DTS transmitters must be synchronized and the clock used for this purpose is the GPS signal.

Various federal agencies are deeply concerned over this potential jamming of GPS. DoD, FAA, & NIST have analyzed this threat and are on record that GPS would be endangered if the FCC proceeds with the final step authorizing terrestrial use of this band of frequencies next to the GPS Band. Laboratory and field testing has been carried out at the request of the FCC, which extended the deadline to Lightsquared to deliver the final test results to July 1.


My colleagues and I have learned a lot about interference between broadband signals into DTV receivers through our experiments in my laboratory. This knowledge led us to simulate the situation where extremely sensitive wideband receivers are overloaded by signals such as those proposed for these 40,000 base stations.

We have carried out those tests, not with LTE signals of 5 MHz bandwidth, but with our ATSC signal of 5.38 MHz bandwidth. As the LTE signals would be upgraded to 10 MHz later in the implementation plan, we used pairs of our 5.38 MHz DTV signals to simulate the 10 MHz LTE signal. The test plan called for tests with both 5 MHz and 10 MHz LTE signals, but last month, we learned that the 10 MHz tests could not be done within the timeframe demanded by the FCC. We have simulated all three phases and our results contradict the notion that tests with two 5 MHz signals is the most critical and therefore tests with two 10 MHz LTE were dropped.

In our simulation, we used DTV signals in the UHF band, not LTE signals in the L band (1525-1559 MHz. By subtracting 959.31 MHz from each frequency in the LightSquared signals, we shifted them to the UHF band.

Fig. 1 shows one 5.38 MHz DTV signal with 3rd order distortion products falling in both first adjacent channels. This simulated the Phase Zero LightSquared Signal. This LTE signal would be from 1550.2 MHz to 1555.2 MHz. The 3rd order distortions products extend from 5.38 MHz below 1550.2 MHz to 5.38 MHz above 1555.2 MHz. None of these come near the GPS Band 1570 – 1580 MHz. However there is another 3rd order distortion product, de-sensitization and this would reduce the GPS signal power thus causing jamming. That cannot be simulated.

“Simulated Phase Zero Signal”. Add 959.31 MHz to each frequency shown to get the LightSquared Signal frequencies.

Figure 2 shows two 5.38 MHz DTV signals simulating the LightSquared Phase One Signal. These two signals can produce 3rd order distortion products, some of which fall in the GPS band and could jam GPS reception. The highest frequency 3rd order distortion product is centered Marker # 4 at 1577 MHz blanketing the GPS Band. LightSquared stated that their Phase One signal would cause the most GPS jamming, if any occurred.

“Simulated Phase One Signal” Marker # 4 corresponds to 1577 MHz in the GPS Band.

Fig. 3 shows the Phase Two Signal. Here there are contiguous pairs of 5.38 MHz DTV signals to simulate the 10 MHz wide LightSquared LTE signals. The spectrum of 3rd order distortion products extends up to Marker # 4, 1584 MHz well above the GPS band (1565 – 1580 MHz). There is a considerable noise power density at the GPS carrier frequency of 1575.42 MHz . This high noise power density might prevent GPS receivers from synchronizing.

“Simulated Phase Two Signal” Note that the 3rd order distortion products extend above the GPB band in both Figs. 2 & 3.

Between markers #2 and #3 in Fig. 3, you should note that the noise floor increased by almost 20 dB. This spectrum, 9.2 MHz wide, in the MSS band may not be suitable for space-to-earth transmissions because of this elevation in its noise floor due to 3rd order distortion products generated in receivers near one or more LightSquared towers.

My simulations indicate the spectrum which may be generated in GPS receivers by LightSquared signals. What cannot be shown by such simulations is whether jamming of GPS receivers results from de-sensitization of the receiver or by third order distortion products or by strong signals passing through the receiver because of their finite selectivity. That question may not have been answered in the LightSquared tests conducted recently and whose results were not available at the time this was written.

Next month I intend to comment on what those tests have accomplished.

Post New Comment
If you are already a member, or would like to receive email alerts as new comments are
made, please login or register.

Enter the code shown above:

(Note: If you cannot read the numbers in the above
image, reload the page to generate a new one.)

Posted by: Brian Smith
Fri, 07-01-2011 11:48 AM Report Comment
Great article! Makes very clear why what seemed to be a cut and dry is actually not so. My only problem is with the headline. SPECTRA! Why did singular forms "datum" and "medium" get bulldozed by "data" and "media" (e.g. "data is..." and "media is...") while "spectrum" seems to have body slammed "spectra"? At least my spell-checker still doesn't think I'm a recta.
Posted by: Brian Smith
Thu, 06-30-2011 05:24 PM Report Comment
Thank you Mr. Rhodes for your continued testing and vigilence in regards to keeping tabs on the FCC and how the RF spectrum is utilized. We can only hope that there are a few engineers left at the FCC who aren't also employed by the wireless lobby that also possess competent skills in mathematics. If they think US TV broadcasters all have a 1000ft tower in a Kansas corn field to determine interference (Longley-Rice) one can only imagine how the FCC will handle this one! Eagerly awaiting your next installment.
Posted by: Brian Smith
Thu, 06-30-2011 02:50 PM Report Comment
As I've often said, "one test trumps 1000 consultants." Testing is a great way to unequivocally determine the effects due to an unknown input or situation. I would take exception to the comment that intermoduation products cannot be simulated. We have, in fact, both calculated and simulated the effects of intermodulation products through non-linear amplifiers. Our analysis requires two simple assumptions; 1)the input waveform is Gaussian-like (as LTE carriers most certainly are), and 2) the amplifier is memoryless and can be described in terms or AM-AM and AM-PM curves using a single CW tone. We have performed such calculations on the LightSquared situation, and have posted those results on our blog ( We will be updating some of those results as our analysis matures. We are simulating a multiple LightSquared tower situation with the GPS satellite constellation using our Wireless Network Simulator (WiNS).

Thursday 11:07 AM
The Best Deconstruction of a 4K Shoot You'll Ever Read
With higher resolutions and larger HD screens, wide shots using very wide lenses can be a problem because they allow viewers to see that infinity doesn’t quite resolve into perfect sharpness.

Featured Articles
Discover TV Technology