PARIS — Dutch IoT/M2M satellite connectivity startup Magnitude Space has signed an MoU with mobile satellite services provider Iridium Communications to investigate how their complementary constellations might work together.
Magnitude expects to start commercial service in 2018 with just one or two UHF-band satellites in low Earth orbit, and then scaling production in 2019 for a constellation that grows only as fast as business demands.
Founded in 2016 and now with just 15 employees, Magnitude got an early lift from the European Space Agency (ESA), whose Business Incubation Centers offer co-financing for promising initiatives.
Magnitude Chief Executive Officer Ernst Peter Having discussed the company’s status and outlook.
Where are you on system design and satellite production?
Last year we completed full system architecture design, then we finalized the full simulation based on the protocol we developed and the modulation technique. We are working towards a launch of our first satellite in early 2018 with about 15 pilot customers.
We have 10 more prospective customers in a holding pattern because we cannot develop hardware concurrently with all of them. They are ready to roll out pilot programs in Q2 of next year, and commercial launch will be in Q2 of next year.
And the second satellite will go up at the end of Q2 next year and that will give us service revenue.
You can start commercial service with just one or two satellites?
Yes, because basically the service level we are offering at launch is sending one message per day. With a constellation with one satellite, you already have two revisit moments per day in the worst case, at the equator.
Your optimal architecture is 48 6-kilogram satellites?
We have a very scalable model. Our analysis has shown that in IoT connectivity, real time is not relevant. Two-way is often not relevant. Most data is used for analysis later on.
So we will grow the network as our subscriber base grows. It’s all about affordability.
We estimate that with 16 satellites we will be able to provide an hourly service. Only as our subscriber base grows and we fill up the satellites will we add capacity. We need to make sure the service is, first and foremost, affordable.
I don’t want to go in the direction of OneWeb or [IoT startup] Helios Wire or some of the other constellations. We have a pragmatic approach that is all about low power, affordability and ease of use.
This is what customers are looking for. They are not looking for a 15-minute revisit time at the equator. With a few satellites I basically have a service in North America, Europe and Australia that is almost every half hour.
You are still formally in ESA incubation until 2019?
I would say that is semantics. We have already raised well over 3 million euros in funding and have got government endowments which are over 1.5 million euros, so we are funded through launch already.
Launch of what?
The first satellites.
The first two?
For the second satellite, we are now working on a Series A that we are going to close at the end of this year. We are funded through the first satellite and operations until close to the end of 2018.
For the second satellite and the follow-on constellation we are currently working on a Series A.
You had raised 1 million euros last year and then were looking for a 5 million euro A round, no?
We postponed the A round and our initial investors, angels, provided additional funding because we found this a better way, so we now have taken all the risk out of the first satellite already.
The first satellite we now have fully funded and we have an in-orbit-delivery contract with our supplier, which takes most of the risk out of the A round. Basically we did an extra bridge round.
You’ve partnered with Hyperion and Integrated Solutions in Space, ISIS. ISIS is a launch integrator, and Hyperion does satellite components?
Hyperion is building our payload and ISIS is integrating the satellites. So they are the prime contractor using, among others, Hyperion components for the payload — the most important component for us because it’s processing the data.
You had a launch at the end of this year. That’s been delayed?
We had a [Indian] PSLV launch originally in November that was pushed out to March of 2018 and then we looked at a different PSLV launch that got us to a better orbital plane, for April. The launch is still scheduled despite the [recent PSLV failing-separation] mishap. So we are looking at options, but targeting a launch data the end of Q1.
Do you have options for the launch if the PSLV schedule is disrupted?
Yes, there are options. We are being prudent and thinking about a Plan B or C because our customers are ready for launch. Our modems are basically being delivered to our customers in October — next month.
So they are finalizing their hardware designs for their applications so we also need to deliver as promised to our customers. So that’s our target.
How big is your modem?
It’s roughly three by five centimeters.
Who builds it, and who has the IP?
The design is a co-production between Hyperion and ourselves, in-house, and we are having it manufactured here in the Netherlands.
Your manufacturing contract scales in 2018?
In the end we are not a hardware manufacturer, we are a connectivity provider. We are in talks with the OEMs that are building applications that in the end can also build these modems directly into the devices as a licensed integrator. Our business model is strictly connectivity. We sell subscriptions.
How low can you get the cost, at what production-run level?
Already, and with the current low volumes, we are well below $50. The aim is to get the modem to below $20.
So below $50 with an initial volume that’s in the hundreds?
The initial customers to our prepaid service and we have a minimum of 500 subscriptions next year, with the initial modem run of about 1,000. There will be others after of course. And even at these low volumes we can get the cost down to this level.
It seems a new IoT/M2M constellation is announced every 15 minutes. What’s your unique selling proposition, the modem?
There are a couple of elements. The protocol that we have developed is really efficient in the number of bits per hertz per second you can actually put through. It’s also the energy effectiveness of the modem, so it’s about the algorithm in the modem that allows for a very effective sleep-and-wake mode.
So while it’s a one-way system from a customer perspective it’s a two-way system from a network management perspective.
That way we can keep the power consumption low. And as we focus on a simplex solution, we can deliver a service with a low number of satellites, which means we keep it affordable.
Beyond that, we have got our [ITU frequency regulation] filings in order, unlike some of the others. We have filed for our UHF frequency and in other bands, and we are not dependent on others.
There is a Swiss-American company that hasn’t even filed yet with the ITU and is hoping for an agreement with one of the geostationary players in L-band, to piggyback. In order to do that they need to get their own ground-to-LEO filing in place. Otherwise their service is not allowed by the ITU.
Making sure you have the regulatory piece in place, in the right bands, is important to make sure you can deliver an affordable low-power solution. Because the moment you go to the higher bands, you can still have an omnidirectional antenna on the ground but you will need to have a more-expensive antenna on the satellite, which will make the service more costly, with low volumes, in the couple of hundred thousands.
There are constellations out there that claim their business case is 6 billion devices. I don’t believe those business cases.
You mean getting to that scale?
Getting to that scale, and getting there soon enough and profitable enough. It’s basically the Iridium trap, an over-designed system for low volumes. Our system is commercially viable even at low volumes, so there is a better chance for our business model to work and get to break even sooner.
The moment you start relying on intersatellite links you need a larger number of satellites to deliver a minimum service level, then there are more risks. We have a low-risk model and we have the technical advantage in that we have a closing link budget with a protocol we developed ourselves that is delivering the throughput in all the simulations we do.
And we have our spectrum. That is a powerful combination.
One more thing: We don’t see ourselves as a satellite company, but as a 21st Century digital company.
Everybody says that these days. Being a satellite company is considered uncool.
We need to make sure the satellite works and that the network works, but half our team has a background in on-line platforms, on-line marketplaces. And I know there are pretty good systems out there being used by satellite companies, but I still think that’s the area where you can make a difference compared to existing players.
From the start we said we need a simple fit-for-purpose satellite constellation or connectivity platform, and then we make sure we have the right on-line tools to make sure we are the easiest to work with from a customer perspective. Because that, in the end, reduces cost of ownership.
You worked with RPC Telecom on the ITU filings?
Yes, it’s a Dutch filing.
Did you get assistance from the Dutch military on the UHF access?
We are in talks with the Dutch military on possible use cases, but not the spectrum. We have been working with the Dutch Telecommunications Agency on the spectrum side.
Your modem has a 10-year battery life?
That estimate is based on sending 1,250 bits of data per day. And a limited number of required locations. Of course, if you want to send much more data than battery life will be different. But we did a comparison of average per-day amounts of data compared to other systems. If you look at the amount of power consumption per unit of data sent, then we have similar power consumption to Loral SX. and it’s 10 times more efficient than other LEO operators like Globalstar and Iridium.
That’s probably the comparison that’s important: power consumed per amount of data sent.
Our calculations have been made using off-the-shelf, standard batteries.
How does the modem know when the satellite is passing overhead?
The modem knows where it is and what time it is. There’s an updated chart in its memory of the satellite constellation so it knows when the next satellite will pass. It starts listening shortly before the satellite passes, the satellite lets it know it’s passing, and so we have optimized the pattern of sending in the population for that pass.
What does Iridium do for you in this MoU?
We are extremely complementary. They recognize that their constellation they cannot deliver a service at the same power levels as we do. In that respect we deliver a service to another range of devices. However, those devices can often be found in the same customer segment as Iridium’s currently serving.
Look at heavy industries, where customers may use Iridium on larger devices but are not using Iridium on their generators and their tiny machines. They would like to have those connected to a monitor as well. But they want a low-power, affordable device.
The same thing goes for trains. The engine will have Iridium, the rail cars need a low-power solution. The moment they are in the yard, they need to make sure that the battery lasts for a long time. So they see we can deliver a complementary service within their customer base and increased commercial opportunity.
You want to make sure you span the whole spectrum. Iridium is in the middle, we are to the left of them and there will be higher-end solutions to their right. Iridium also has expertise we can benefit from.
And we have expertise because we are in touch with a lot of terrestrial operators, and terrestrial IoT connectivity providers looking to us for a low-end satellite solution, and basically we can bring in Iridium to move up the scale.
What’s the end state of your Iridium relationship? How far will it go?
We have a memorandum of understanding that leads to a full agreement toward the end of this year. We are working out the details on how we will fill in the partnership. We both see benefits but at this point it is a memorandum of understanding.
It seems the agreement makes no sense unless it goes much further.
The problem with existing satellite constellations is that they are always lagging behind. The beauty of a system like ours is that I write off my satellites in three years. The time to market from when we freeze the technology give an order to ISIS, have it produced and then launched, is about six months.
I am refreshing my constellation every three years. The 10 satellites that are going to go up in 2019 will be replaced in 2022. And the satellites going up in 2020 are going to be more advanced. It’s really a software-defined radio.
This means we can be much closer to developments in the terrestrial market. Existing players may see an interest in working with us at arm’s length, to see what’s happening in our realm without taking the risk, and they can deliver an Iridium-type service under their flag and do price segmentation. They deliver to the high-end customers, we deliver to the low-end customers. We are the low-tech solution.
M2M/IoT is Iridium’s fastest-growing business.
We believe that by having end to end solutions we can make the pie bigger. It’s all about growing the total pie — their piece and our piece.