Satellite 2021 – LEO Digital Forum: What New Things Did I Learn?

Satellite industry’s flagship annual conference “Satellite 2021” kicked off last week with the “LEO Digital Forum”. Mark Holmes and his team @ Via Satellite did an excellent job in gathering LEOS world’s Who’s Who, with Elon Musk (and Amazon Kuiper, for that matter) being the only notable absence. However, SpaceX was well covered by its COO Gwynne Shotwell, who also won the Satellite Executive Award for 2020.

Personally, this has been fascinating to track the LEOS space for over 3 years now. For sure, there is no short supply of new learnings to confirm or update my view from what I said on this in early 2020. Below goes the status update on key LEOS players from my last iteration in February, 2021. 

As this is a fast-moving space, a lot more have happened over the past 2 months. Most notably, as of April 7, SpaceX successfully completed their 10th launch of the year with 1,378 satellites in orbit. OneWeb too sent another batch of 36 satellites on March 25, bringing its total in-orbit constellation to 146 satellites.

However, I want to highlight the following from what I heard firsthand at last week’s LEO Digital Forum-

1)     Gwynne Shotwell, COO @ SpaceX Starlink

·        In 5 years from now, Starlink’s addressable market can be all of rural America’s 20M households or 60M people.

·        There won’t be any tiered offering. Instead, the current plan is to offer a single speed tier for all Starlink users in all markets.

·        Customer Premise Equipment (CPE) cost per unit was ~$3K to begin with. That cost now has come down to close to half of that. The target is to bring that cost down to a few hundred dollars over the next couple of years.

·        Acknowledgement that they still experience intermittent service drops and that’s why they continue to call the current service “Beta”. They do NOT have a clear timeline yet on when they can make the network more reliable to transition to “Regular” service.

·        Reiteration that their (LEOS, in general) service is best suited for rural and rural-ish areas, as a complement to big Cable MSOs/Telcos. The focus is not in offering huge bandwidth in markets well served by incumbent terrestrial providers. Rather, they focus on serving currently unserved (and underserved) markets.

·        To keep costs within reach, they will keep making improvements in satellite manufacturing and launch technologies. (And, of course in CPEs)

·        Check here for more on Gwynne Shotwell’s view on the potential market size and how it differs with other industry insiders.

2)     Neil Masterson, CEO @ OneWeb (Apparently this was Neil’s first public address after taking the top job)

·        Current (Gen 1) constellation of 648 satellites will be deployed by end of June 2022 with a global coverage target by end of 2022.

·        OneWeb will have 40/41 ground stations worldwide. This is significantly fewer than the ground stations count (expected to be few hundreds) for Starlink. The count is dictated by constellation size and design (including altitude).

·        Higher altitude (~1,200km) and resulting wider coverage/satellite means that they will not require Inter Satellite Links (ISL) in their satellites, at least not in Gen 1.  

·        OneWeb’s focus will remain on Government, Aviation, Maritime, and Telco verticals. In terms of telco offerings/partnerships, they will look into Alaska, Indian subcontinent, Africa, and the home territories in UK/EU.

·        (In response to a question that I asked during the Live Q&A session) Neil thinks that 3-4 LEOS players per market is sustainable, partly due to some difference in their respective focus areas. Neil acknowledged that this view is in slight contrast with the view expressed by OneWeb chairman Sunil Bharti Mittal who earlier mentioned that it will be difficult to have more than 2-3 LEOS players/market.

·        More on Neil’s interview can be found here

3)     Dan Goldberg, CEO @ Telesat

·        Once the full constellation is deployed, users will see 6 of their satellites (all equipped with phased array antennas) from any point on earth.

·        Out of their total capex budget (~$5B), 40% will be funded by equity and the remaining 60% will be funded by debt. Also, they will be “Done” in the next couple of months as far as financing in concerned.

·        Expectation is to be fully global by end of 2024. By then they will have 50cm CPEs.

·        As launch partner, Blue Origin has been confirmed. There will be more announcements on more partners in the coming months.

To be clear, none of these findings is radically new. There is nothing there to alter my assessment on LEOS potential, constraints (e.g. localized network capacity), competitiveness and business case. Still, it’s always helpful to hear from the leaders themselves to confirm and re-calibrate the assessment on an ongoing basis. (My assessment has been developed and shared widely with CableLabs members and is outside the scope of this blog post)

Time to Start Paying Attention to CurvaNet?

Personally, I found it more interesting to hear from the CEO of a much less talked about peer of the 4 LEOS mainstays (above 3 + Amazon Kuiper). Tom Choi and his CurvaNet venture first came to my radar when I was working on a LEOS strategy brief/whitepaper for CableLabs members over the summer of 2020. I took note when this previously outspoken LEO critic detailed his plan for yet another constellation, CurvaNet. Below goes my key takeaways of what I heard from Tom at the LEO Digital Forum (with full disclosure, I approached him following his talk and he graciously accepted my invitation to connect)-

4)     Tom Choi, Executive Chairman @ AirSpace IX, CurvaNet

·        More than 2,000 Non-GEOS (NGEOS) constellation plans have been filed to date with the hint of $100B investment.

·        Global telecom/connectivity market is worth $1.7T today, with 3.2B people still without internet. So, to bring in another 1B people for potentially another $1T revenue, a $100B investment makes sense (at least, in theory)

·        However, often a key parameter is missing from these discussions and that’s electricity. Globally 1.2B people live without electricity. Phase array antennas are power hungry. That power does not exist in many rural areas. And, even if available, electricity rates are very high in emerging markets. 

·        Also, $1,200/year (based on $100/month currently being charged by Starlink) tariff is NOT sustainable for large scale adoption beyond the US.

·        The ideal LEOS constellation should have 3 key attributes – affordable CPE, sun-powered design (no dependency on utility grid), and attractive tariff.

·        CurvaNet has been cost conscious from inception – their 240 LEOS constellation calls for $600M in capex - less than 10% of what the BIG guys are planning to invest. More importantly, their 5W CPEs are designed to be at $100/unit. According to Tom, these patent pending “Beach Ball” antennas can offer 50-100 Mbps for Downlink and 5-10 Mbps for Uplink. Their gain is not as good as parabolic antennas.

·        For CurvaNet, the tradeoff is to keep the constellation closer (lower altitude) and cheaper. But then target a 95% service availability, which can potentially be extended to 99% by working with GEOS partners. Tariff target is $0.2/GB/Month.

·        Fund raising is expected to happen in H2, 2021 with demo satellites planned for Q4, 2022.

·        Pilot program will be launched in the Philippines in partnership with Globe Telecom, specially to reach 5% of the population who live outside of Globe Telecom’s tower reach.

·        With regard to spectrum/regulatory issues, Tom claimed that they had hired one of the premier regulatory experts in the world (announcement to be made on May 1st). Unlike other LEOS players, they won’t go after Ku and Ka bands. Instead, they will be focusing on terrestrial lower band spectrum that can also be used for satellite communication. For individual markets, they will either work with incumbent spectrum holders or seek local licenses by themselves.

·        More importantly (and this was news to me), they identified the US and India as their top 2 markets, followed by Indonesia. Tom claimed that they are involved in major discussions with partners (telcos, spectrum owners, tower providers) in all of these 3 markets.

·        Tom concluded with a high note saying that if anyone can be successful in LEOS, that should be CurvaNet.

SpaceX’s Starlink program is the undisputed leader in the industry. Amazon Kuiper project is believed to be at least 2 years behind where SpaceX stands today. Whether new challengers such as CurvaNet still has room to flourish is yet to be seen. Regardless, if you are working in the telecommunications industry, it’s hard NOT to be aware of major happenings in this space (thanks, in part, to the involvement of BIG Guns – Elon Musk and Jeff Bezos).

By any measure, LEOS is a fascinating technological feat. I look forward to attending, interacting, sharing and learning more on LEOS from the main events of Satellite 2021 conference, which is scheduled to take place in DC in late July.