Despite a few high-profile efforts from the hyperscalars, “Internet everywhere” remains a stretch goal. Still, how pressing is an immediate need for a multi-player gaming experience while sailing solo across the Atlantic, hiking the Australian outback or in rail freight across the plains of Rajasthan? There’s a lot of useful value to be had in these remote locations through more basic narrow-band connectivity. For asset tracking in containers which may be at sea, on a long-haul train, for fishing vessels, farmers or utility operators, the most obvious way to provide broad coverage is through satellite networks, and very good way to handle low bandwidth traffic is through NB-IoT.
Satellite NB-IoT is coming fast
Satellite-based NB-IoT is already well beyond a concept. Skylo Technologies for example have demonstrated NB-IoT proof of concept systems based on existing geostationary satellites. No enhancement is required to the satellites themselves (naturally). According to the CEO, these act simply as a mirror, connecting to small earth-bound hubs. Skylo claims this solution costs only 5% of what other options will demand, the hubs themselves costing less than $100 a piece, making this truly accessible to everyone. Skylo has over $100M funding from the SoftBank group and is already in partnership with the Indian telecoms company BSNL.
Sateliot is working with the Gatehouse group with a different approach. They plan to deploy a constellation of small satellites which will act as base stations, in low orbit around the globe. Meanwhile OQ Technology has run trials on low-earth orbit satellites and is planning services ranging from third-party satellites offering their SDR-based service to a constellation of their own satellites. OQ Tech said they have already worked through the Doppler and range issues unique to non-stationary satellites. In the US, Ligado Networks has plans to build a satellite network, supporting this technology. There are also hints from China, who launched two IoT satellites in May, as a start to building a larger constellation over the next 2-3 years. This is a very active field, already past proof of concept, now moving into building actual networks.
The need for software flexibility
That said, the current 3GPP standard doesn’t fully support non-terrestrial technology. What these companies have done has worked around those limitations. 3GPP is now working on Release 17 to better support the non-terrestrial use-model. Which raises an obvious question: can such a change be supported through a fully SDM (software defined modem) implementation? If it can’t, pilot deployments are going to have a problem, especially if depending on hardware embedded in an already-launched satellite.
There’s another important consideration. While you need to exchange data with your IoT device, often you also need to know where it is. Consider an asset tracker, on a container, on an ocean-bound carrier. That device should provide GNSS support, the global version of GPS, which ideally should fit onto the same platform, adding little additional cost.
This is a very exciting new market opportunity, an opportunity in the middle of rapid evolution. Some of this evolution will impact not only the satellites but also ground-based implementations.
Edited by Ken Briodagh