The internet of things has a Low-Power Wide-Area Network (LPWAN) problem. Most people probably look at the huge number wireless protocols and companies trying to send small bits of data wirelessly over long distances and think the problem is that there are too many of these companies. After all, there are telcos offering NB-IoT, proprietary efforts such as Sigfox and Ingenu, LoRa networks such as Helium and Senet, and satellite startups such as Astrocast.
But the real problem is economic. As we’ve talked about on the podcast and written about separately, it costs a lot of money to build a tower-based network designed to cover large areas and send inconsequential data. And by inconsequential, I mean that the data is small, but also that — individually — it’s not valuable. This is not data for which a regular consumer is going to pay cellular rates. Rather, it’s designed for products that will sell for just tens of dollars.
The economic challenge is also why we keep seeing new companies trying to get into this space, armed with new ideas for how to reduce the cost of connectivity. So far, Helium’s effort to build a LoRa network using a shared, distributed network is most compelling, although I’m hopeful that some carriers will be able to make NB-IoT work.
This week, I met with Totum Labs, an 18-month-old satellite company that plans to tackle the economic problem facing LPWANs with satellites and a low-cost, custom chip. Ted Myers, the co-founder and CEO of Totum, may be familiar to industry watchers as the former co-founder and chief technical officer at Ingenu, which was trying to build a tower-based LPWAN.
Myers says that, while Ingenu is still around, most of the founding team left once it became clear that it was too expensive to build the network in order to get companies to build devices to run on it. So he’s turning to low-cost nanosats to deliver broadband connections for low-powered devices around the world.
Totum is seeking funding to launch an eventual 25 satellites by 2022, building what will be a worldwide network that can transmit low data-rate messages with the unlicensed 2.4 GHz band used for Wi-Fi when on the ground. The chip that companies will use in their devices will cost $4, which is cheap enough to get it into consumer devices that sell for around $15-$20.
Each satellite can support half a million devices, so the idea is that within a year of the launches the company could support a fully global network for millions of devices as opposed to having to roll out a network geographic region by geographic region. Myers is especially excited because — thanks to the 2.4 GHz spectrum — the Totum technology will also work indoors. The satellite connection will communicate with a gateway that devices connect to. In the meantime, Totum has signed a partnership to have existing company Loft Orbital provide the gateways and ground network that Totum will use for backhaul.
Totum has plenty of competition in the sky. Traditional satellite companies are waking up to the opportunity for IoT, as are startups such as Myriota, Lacuna, and others. But Myers is no stranger to the world of connecting machines, and in March raised $10 million in Series A funding, which will pay for the first satellites to get off the ground.
After that, the business will depend on companies building in Totum’s modules so their devices can talk to the network.
Stacey
Great article expanding on future IOT possibilities in areas not normally empowered with good internet connections. Personally, I am trying to understand how a company like this might be able to compete with Elon Musk’s StarLink, that went Beta last week. Although the receiver is currently larger (about the size of a pizza box) they are expecting speeds of up to a gigabit per second at latencies from 25 milliseconds to 35 milliseconds. This will allow the building out of mesh environments where IOT could thrive with little setup and cost, yet still be secured, segmented, and encrypted without a degradation of bandwidth.
What are your thoughts on StarLinks impact on the IOT sector, especially in rural and rugged environments.
Ian,
Totum is building a thumbnail-sized System-on-Chip (SoC) that has an integrated 2mm antenna so the device size is going to be the size of a domino – or perhaps your thumb if you opt for more battery. It’s purely IoT data and the BOM cost for that modem functionality is roughly $4 USD. Data connectivity plans will be highly competitive in volumes – far below what those high speed IoT offerings would be able to offer.
So if you have a remote facility (well, meter, livestock, etc.) this becomes the easiest, lowest cost option to send/receive alerts. Or, if you are tracking high-value assets you can integrate their SoC into your motherboard to track it like a stolen iPhone or integrate a tiny device into the Pelican case itself…or you have pallets, package deliveries, or anything worth tracking at low intervals (basically anything not real-time) this is a phenomenal breakthrough in discreet size with low cost.
It’s important to note that Stacey has one sentence that’s not accurate in her article. Totum does NOT rely on a local gateway like a WiFi router in your home. The device itself connects to the satellites. It’s a credit to a brilliant and complex algorithm the company developed to make it possible.
If you want to learn more feel free to ask questions here or hit me up directly at eric.daversa [at] totumlabs.com
Hi Ian,
The Key for businesses like Totum and eSATGlobal.com is that we provide direct to satellite connectity. THey are aimed at devices with low data requirements and long battery life requirements. Examples would be tracing a cow in a field or a soil sensor detacting the level of carbon sequestration in the soil.
Starlink require large antennae, the size of a pizza plate, and the power required to reach their satellites would run down the batteries of a small sensor in minutes. Starlink is designed exclusinvely for broad band.
GEO’s still have a number of advantages of LEO’s eSAT is piggy backing off exsiting GEO’s with Inmarsat and Thuraya. These GEO’s have the capacity for 1 billion devices each vs 500,000 for LEO’s and will consequently have subscription charges much lower.