SmartSky Networks, the company that plans to launch a new 4G air-to-ground (ATG) network in the US, continues to forge key agreements with respected companies in the inflight comms space, and even recently snagged its first “Early Bird” customer in the form of private aviation operator DAVINCI Jets. But despite pressure from various parties, including media, SmartSky is holding one card close to its vest – it is not revealing what spectrum it will use, except to its nearest and dearest.
When RGN put the question to SmartSky in advance of the NBAA show in Las Vegas, company president and director Ryan Stone said, “Mary, great question but no, no more specifics on that topic this week. We are releasing many details related to our Early Bird program, including promotional pricing for hardware and service.”
Naturally, if SmartSky was a publicly traded company, its lack of transparency would not be tolerated. But it’s not public, so its decision to remain quiet on spectrum for competitive reasons – though frustrating and driving a level of skepticism in industry – is entirely its prerogative. The most it has said is that it’s eyeing a variety of options.
Gogo’s former manager of content partnerships, Mary Rogozinki, as VP, airlines. In this new role, she is responsible for helping define products and business partnerships “to grow SmartSky into an IFEC leader in the airline sector and for leading the continued development and implementation of SmartSky’s business model with airlines, airline distribution partners, and customers”, the firm tells RGN.
SmartSky’s Stone says he is excited to have Rogozinki join the SmartSky executive team, adding that she “brings extensive experience in the IFEC industry, including her deep involvement and leadership of the Airline Passenger Experience Association (APEX).” Indeed, before working for Gogo, Rogozinki was a long-time executive for United Airlines. She is currently the co-chair of the APEX Education Committee, and has close ties to Hollywood’s non-theatrical executives.
But Rogozinki is at no more liberty to divulge SmartSky’s spectrum plan to media than is Stone and so we’re left to guess again. Donning the journalist cap, long-time industry consultant Peter Lemme has penned the following analysis, “Smartsky – left to guess?” for us (it differs from his assessment last year when SmartSky first came on the scene in a more public way). Feedback is welcome. For its part, Gogo – the dominant provider of connectivity to bizjets flying domestically – continues to dismiss SmartSky as being any sort of serious threat.
Smartsky – left to guess?
Smartsky has continually declined to reveal their fundamental radio technology, rather stating:
- the technology is fundamentally 4G LTE
- they have 60 MHz
- they have successful flight test demonstrations
- they have survived diligence from investors and distribution partners
- Trust them, it works.
Skysmart have declined to identify an FCC license, or make any comment as to the source of the spectrum providing their radio link. Smartsky has no reason to deny an FCC ruling or license, as that is a matter of public record.
I appreciate that Smartsky investors, suppliers, partners, and even customers are convinced this technology is viable. I appreciate that any technology has its skeptics. What I don’t appreciate is having no way to judge, to be left to guess, and to worry about what is being hidden. I welcome a trial as an example of a working technology. A reliable airborne communications network creates a myriad of challenges and operating conditions typically discounted in a trial; it is no small step!
A barrier to entry from competition is the greatest boon to any investment. Holding sensitive information confidential creates a challenge to compete. I would never suggest a company reveal their special sauce just to satisfy curiosity.
Aviation communications represents a significant investment and commitment. Service providers have come and gone, and that can leave a legacy of airplanes with useless equipment and no service. The tenants of any prospective service are viable technology, spectrum, and sufficient funding. As a technical commentator, I am committed to understanding each service provider well enough to be convinced how it works, how it will scale, what are the risks to its capabilities, and how does it fit into the marketplace.
If Smartsky is viable, with no license, then what spectrum makes most sense?
ISM 2.4 GHz band? This band is unlicensed, which means not needing a radio license as long as you play by the rules, and putting up with everyone else using the band similarly.
In my analysis (see below) which is quite cursory, hardly definitive or authoritative, and possibly seriously flawed or entirely irrelevant, everything looks to align with Smartsky claims. The regulations appear to be supportive, except for the issue of use against a mobile (moving) target instead of a fixed (not moving) target.
Smartsky could plead for an FCC waiver for mobile use of a high gain antenna in an ISM 2.4 GHZ unlicensed, airborne, point-point LTE application spanning 60 MHz and two polarizations.
I have no certainty any of this, that LTE is not my specialty, and that I have many questions unanswered. Yet, it seems to be conceivable. I wish Smartsky would address the skeptics with data and analysis. Stand up to their claims and bear scrutiny to build confidence, rather than hide behind confidentiality, which builds suspicion.
Unlicensed ISM 2.4 GHz LTE Line of Site (LOS)
Assuming an unlicensed frequency band was a bit like Goldilocks, with 60 MHz the temperature. 900 MHz too cold (small), 5.8 GHz too hot (big), 2.4 GHz just right.
Channel 1 starts at about 2401 MHz and Channel 11 ends at about 2473 MHz, a span of about 72 MHz. Frequency spacing of a carrier is around 1.2 typically, so a 72 MHz allocation yields about 60 Mhz of usable carrier space and 12 MHz of guard band. This could be further subdivided into smaller carriers and associated guard bands.
Using unlicensed spectrum for LTE applications is being pursued in the cellular industry, but not in the 2.4 GHz band, and not for airplanes – rather for short-distances as a gap filler.
Unlicensed LTE is a protocol to allow concurrent operations with Wi-Fi, whereby the LTE interfering signals dodge the Wi-Fi signals.
I am assuming Smartsky is building a network that won’t interfere with Wi-Fi by virtue of full immunity with adjacent Wi-Fi users. LTE demodulators typically can run about 10 dB lower (carrier over noise plus interference levels) than compared to Wi-Fi demodulators. The lower LTE signal levels might reduce the impact of unwanted LTE emissions upon Wi-Fi, but the Wi-Fi signals present a source of interference to LTE too.
Unlicensed LTE on 2.4 GHz for an airborne terminal has different challenges, but it brings with it some advantages too.
The candidate strategy would rely on:
- a high gain antenna at the base station
- that the airplane does not transmit below some threshold (10,000 feet) above the ground.
Numerous examples of a terrestrial link to airplanes use a cross-pole antenna on the belly of the airplane. Two antennas would permit a simultaneous connection to two base stations. The cross-pole operation offers a second spatial stream, or effectively two 60 MHz carriers could be used from one base station utilizing both polarizations.
In the ground to air direction, the base station emissions would be focused at the airborne target, well above the surroundings. By careful shielding and the use of a high gain antenna pointed into the sky (at an airplane), the base station emissions might not couple to the surrounding ISM users, and avoid interference.
Similarly, the base station directivity might offer some immunity from the surrounding noise created by proximate public usage in the ISM band.
A typical line-of-site antenna has a gain of 6 dBi or less, which is within regulations for the unlicensed band. I would assume a total EIRP no greater than a typical access point, again to remain within unlicensed emission regulations. Airborne emissions would illuminate the surrounding community, separated by a minimum altitude. I can only assume Smartsky believes the airplane emissions would not interfere to those overflown, due to the very low interference level it would present, and by remaining within regulatory emission limits.
I can only assume Smartsky has a plan for creating point-point beams and tracking airplanes. The overall capacity of a base station is predicated on how directive the beam. It is not apparent whether more than one airplane operates in a given beam, which would cause frequency/polarization sharing, and possibly complicate compliance findings.
Using a high gain antenna is an option.
Code of Federal Regulations, Title 47, Chapter 1, Subchapter A, Part 15, 15-247
Operation within the bands 902-928 MHz, 2400-2483.5 MHz, and 5725-5850 MHz:
(i) Systems operating in the 2400– 2483.5 MHz band that are used exclusively for fixed, point-to-point operations may employ transmitting antennas with directional gain greater than 6 dBi provided the maximum peak output power of the intentional radiator is reduced by 1 dB for every 3 dB that the directional gain of the antenna exceeds 6 dBi.
15-247 refers to a fixed service, this is a mobile application. Smartsky would need to plead that the ground to air transmissions are sufficiently directed above any terrestrial adjacent user such that all adjacent users are immune.
I think these pleadings could be resolved by a waiver or some ruling from the FCC, after a thorough analysis with measurements and public commentary. Not sure the stand of any Wi-Fi trade/alliance, but presumably the credible proof of no interference should placate those concerned.
Smartsky may not need to plead that the air-to-ground transmissions are weakened sufficiently by the criteria of being above a minimum height over the ground to not interfere to those over-flown (while the base station is super-sensitive).
Another question is how to use the two 60 MHz paths to and from each airplane. By using a highly directive base station antenna, each beam can re-use spectrum. Cross-pole isolation may be a factor with adjacent beams, or further frequency subdivisions may be necessary. I will assume only one polarization is in use.
The full spectrum is usable in both ground to air and air to ground. The signal will scale with distance, and presumably the data rates as well. Mobile operations, particularly maneuvering, would require rapid response to fading, and robust margins above threshold.
The ground to air receiver is picking up a lot of surrounding noise which may diminish the spectral efficiency. Basic processing would suggest at least 0.1 is conceivable, so at least 6 Mbps seems in reach, with the potential for much higher rates as the airplane operates more favorably to the base station.
The air to ground signal seems most in jeopardy at long range, very dependent upon the receive antenna gain to both create immunity from surrounding noise and to close the link. A similar ground to air antenna gain may be needed to close the link. Higher gain in both directions would reduce the resultant beam size. The broad wireless industry is hotly pursuing how to create and use spatial streams. I am beyond my understandings here and, but the technology seems to be at hand. My napkin, however, is in tatters.