In a recent conversation with Peter Lemme, analyst and chairman of the Airlines Electronic Engineering Committee (AEEC) Ku/Ka-band Satcom Subcommittee, RGN picked up an interesting bit of intelligence that warranted further investigation. The tip concerned Qualcomm’s proposal, currently under consideration by the FCC, requesting that 500 MHz of spectrum in the 14.0-14.5 GHz frequency range be auctioned off in shares to companies that want to offer inflight connectivity via air-to-ground (ATG) networks.
In the eyes of the FCC and according to law, terrestrial networks are a secondary service which must always yield to primary satellite communications services; the 14.0-14.5 GHz is in the Ku-band of spectrum. Qualcomm of course is well aware of this and seems to have demonstrated in its proposal that it can avoid interfering with geostationary satellites by ensuring that all of its future US-based ground stations transmit exclusively northwards – or away from the geostationary arc. That’s some good news for Qualcomm… but not all satellites can be avoided by simply pointing your beam to the north.
Lemme posed the question of what might happen should Google’s ‘WorldVu’ Ku-band NGSO constellation – aimed at providing broadband access to communities around the globe – make it into orbit. “I think the Qualcomm initiative, which wants to re-use the 500 MHz of Ku spectrum, is predicated on the fact that there is no NGSO constellation in Ku-band,” Lemme revealed in that discourse. “There is already [a constellation] in Ka-band and that is 03b, but I think the Google initiative, if realized, will undermine the Qualcomm initiative completely. I haven’t seen anybody make that connection.”
Lemme was onto something.
If all goes to plan, as suggested in multiple reports, the first half of Google’s Ku-band WorldVu constellation will enter non-geostationary orbit (NGSO), in this case ‘low earth’ orbit, in 2019. NGSO satellites, such as Iridium’s network of 66 active units positioned about 780 kilometers from the earth’s surface, move through the sky at a very rapid pace, not matching the earth’s own rotation like geostationary birds. The WorldVu constellation is expected to feature 360 satellites spread over two planes, one at 800 kilometers and one at 950 kilometers above sea level. If launched, a Google satellite could potentially cross the North American horizon once every eight to ten minutes, using the same 14-14.5 GHz frequency as Qualcomm’s proposed terrestrial network.
Assuming that the FCC goes through with what Qualcomm is proposing and allows the 14-14.5 GHz swath of Ku-band spectrum to be re-used for ATG services – and it does appear to be leaning in the direction of an auction – the onus will always be on the terrestrial network to avoid alignment and interference with any future satellite networks. This applies even if Qualcomm or whomever wins licenses at auction launches said ATG ahead of Google.
We called on industry analyst and President of TMF Associates, Tim Farrar, to see what he could make of all this.
“It is highly likely that there would be alignments very, very frequently,” he said. “The challenge is that the uplink transmissions from the ATG gateways would have to avoid illuminating any NGSO satellites … [an ATG network] would have to turn on and off every time the link from the ATG station to the plane came in line with an NGSO satellite going past. I think it is therefore pretty clear that the Qualcomm proposal would be extremely difficult to execute if WorldVu decided to offer coverage in the US.”
Although Qualcomm told RGN that it is “not in a position to comment” on Google’s plans, the company has assured the FCC that any future NGSO operation will have priority over air-ground mobile broadband. Yet Farrar pointed us towards some evidence that the communications giant is anticipating problems. An FCC filing submitted by the SIA (Satellite Industry Association) on May 21, 2014, suggests that Qualcomm has recommended ‘coordination’ between primary satellite services and its proposed terrestrial structure. The SIA is not keen on the idea.
“ … SIA objects to Qualcomm’s proposal that future NGSO FSS systems should have to “coordinate” with secondary Aeronautical Mobile Service (AMS) operations, implying that the NGSO system may need to accommodate the secondary AMS in some way. If Qualcomm’s AMS is truly a secondary service, it must not constrain existing or future primary uses.” The filing goes on to argue: “Conversely, as a primary service, NGSO FSS systems should be under no obligation to coordinate with secondary services, whether pre-existing or not. At most, a future NGSO FSS system should only be required to give notice that it is about to commence operations and provide a description of the nature of such operations.”
So what does all this mean for Qualcomm’s proposal and the FCC’s pending decision to auction the spectrum or not?
The FCC could still choose to auction the spectrum as Qualcomm is requesting – something Gogo is actively supporting, though it disagrees with specifics regarding how the spectrum licenses should be auctioned. And, as Lemme notes, “It appears that there may be a path forward to manage interference to NGSO using a similar method as used towards the GSO (power spectral density spectral mask), but I cannot assess the levels that would be chosen as being restrictive or liberating. In any case, NGSO license holder has the primary rights while suitably separated from the GSO, and any applicant would need to prevent interference towards the NGSO network, while accepting any interference from the NGSO network.”
Yet, the attractiveness of owning that spectrum, along with its sale price, are likely to diminish, suggests Farrar. Assuming the spectrum is sold, Google might decide to play nice with future ATG operators sharing their spectrum after all, but they certainly don’t need to. “Fundamentally, if they [Google] have primary rights to the spectrum, it certainly is a problem. Unless they were cooperative, they would have the potential to block the use of that spectrum for the ATG network,” he says.