Hardware to Heaven: Overcoming the Challenge of Proprietary Connectivity Hardware in the Satellite IoT Market

Subscribe To Download This Insight

By Tancred Taylor | 4Q 2022 | IN-6698

Announcements about satellite Internet of Things (IoT) have proliferated this year. As satellites become easier to launch, satellite operators have been held back by proprietary hardware. An open hardware ecosystem is a key area of focus, with features, cost, and component availability priorities as the satellite IoT market continues to develop.

Registered users can unlock up to five pieces of premium content each month.

Log in or register to unlock this Insight.


News, and Lots of It


Since the beginning of 2022, there has been a raft of announcements around satellite Internet of Things (IoT). These can broadly be divided into three categories. First, announcements concerning satellite launches or commencement of satellite network services; second, announcements about satellite connectivity partnerships with Mobile Virtual Network Operators (MVNOs) and traditional network operators, showing how the technology will be commercialized; and third, announcements concerning the device-side connectivity hardware, particularly by satellite network operators offering a service based on a proprietary satellite communication protocol, but also by network operators aiming to use existing terrestrial protocols (LoRaWAN, Narrowband IoT (NB-IoT)). The former shows the growing availability of satellite network service for the IoT, while the latter shows the catch-up effort to produce end devices actually capable of communicating with the satellite network.

In February 2022, Orca Systems announced the completed design of its ORC3990 System-on-Chip (SoC), designed for Totum Labs’ satellite constellation. The SoC operates using Totum Labs’ proprietary protocol on 2.4 Gigahertz (GHz) and 5 GHz unlicensed spectrum; the use of unlicensed spectrum provides Orca Systems and Totum Labs with a lot of bandwidth, an important factor in increasing network capacity and simplifying device design because devices using these frequencies on the SoC can be built with a simple Printed Circuit Board (PCB) trace or Surface-Mount Technology (SMT) chip antenna—a significant advantage over many competitor solutions operating in the sub-GHz spectrum bands. The SoC integrates Totum Labs’ proprietary modem, for which Orca Systems is currently the only fabless semiconductor company to have a license. The system allows for a 10-year battery life (based on two AA batteries), sub-US$10 end node price point (for smart label type devices), and indoor Sensor-to-Satellite (S2S) connectivity. Orca Systems’ plans were for 4Q production and launch, with the goal of starting to service the backlog of 2 million advance orders.

In June 2022, chip-to-cloud technology provider and ZETA protocol owner ZiFiSense announced ongoing work with French satellite operator Kinéis on an SoC. Kinéis operates a proprietary, lightweight, and low-power communication protocol. The SoC with ZiFiSense will be software-configurable, supporting on the same chip two radios and two basebands allowing the SoC to connect both with terrestrial ZETA gateways and with Kinéis’ overhead satellites with different modulation formats (M-FSK for ZETA, and BPSK for Kinéis). On September 14, 2022, Kinéis further announced its entry in the STMicroelectronics (STMicro) partner program. According to the announcement, this will allow developers to use STMicro’s well-used STM32WL sub-GHz wireless SoC to build with the Kinéis protocol stack. STMicro is putting considerable investment into growing its STM32WL SoC availability, based as it is on Semtech’s SX126x transceiver; STM has seen considerable success in this product line since licensing the technology from Semtech. The ability for developers to build on a commonly-used and readily-available off-the-shelf component will allow Kinéis to broaden its market reach and the flexibility with which its solutions are deployed.

On September 20, 2022, Astrocast announced its collaboration with CEA, a technology organization that developed a new architecture for the chip that will work in Astrocast’s Astronode S module. It is unclear where the novelty lies in this announcement, because Astrocast has been working with CEA for multiple years and already had an Astronode S module available. The announcement could signify a new iteration of the chip that takes into account feedback from trials with Astronode S on the Astrocast network, or it could be a sign that the companies want to publicize their commercialization of the product as a way of generating interest and design wins—in line (somewhat) with Astrocast’s announcement of the commercial launch of its satellite network back in February 2022.

Hardware Is Boring, So Why Talk About It?


The conversation around the merging of satellites with the IoT has been a will-they-won’t-they proposition for the past 5 years at least. The discussion has focused almost exclusively on deploying constellations. Until 2021, companies had consistently fallen behind timelines for deploying constellations or for beginning their network operations. But in the past 2 years, in particular, the satellite IoT market has become extremely crowded simply because of how much easier it is to launch small satellites. As announcements around launches have become more regular, they have become less noteworthy; we can accept that operators are now successfully launching their constellations. These new announcements around the device hardware seem to reflect a shift in the message that network operators want to put in front of potential customers: not only is the network available, but there are now lots of different ways developers can build solutions that work on that network. Satellite operators have been focusing, probably rightly, on the science behind getting their constellations operational; they are now playing catch-up with the commercial aspect by getting modems, chips, and modules ready to be deployed on a large scale.

The move is equally interesting because satellite operators have no long-term interest in making money off hardware sales from chipsets and modules. Even Wyld Networks, a module vendor to the satellite IoT community, isn’t interested in hardware sales: it would happily give away the hardware to customers, or even move to a module licensing model, if it managed to sign contracts for the long term with recurring connectivity revenue. It has, in fact, taken a step in this direction (again in September 2022) through its partnership with Miromico, which will assist in the design and sales of Wyld’s modules.

Generally, satellite network operators have been forced to develop the hardware simply because Space is Hard, and it requires very specific software and hardware configurations to send a message from an IoT end node to a satellite. Operators have had to define the hardware and software on both the satellite and the software on the chip; building chips and modules off the back of this is an effective platform both to test out their constellations and make necessary adjustments, as well as to showcase the technology to the broader market. Currently, however, too many satellite operators’ solutions are held back by proprietary hardware. This is true especially for operators developing proprietary solutions where developers have had to build everything from scratch; but it is also true of operators using standard terrestrial protocols for S2S connectivity. EchoStar Mobile, for instance, offers a proprietary module (using a standard LoRa chipset) for customers to use on its recently launched LoRaWAN-to-satellite Geostationary (GEO) satellite network service.

The announcements are also a bit of a kickback against the domination of satellite IoT headlines recently by those operators using standard terrestrial protocols. NB-IoT and LoRaWAN are the two front runners here, gathering interest precisely because of their ease of use on the hardware front: standard NB-IoT or LoRa chipsets can be used for S2D communication, without any change required on the hardware front at all. The ready availability of chipsets and modules in this market likely makes the proposition presented by these operators a bit more frightening for satellite operators operating a proprietary protocol: whatever advantages they may have in their protocols optimized for low-power S2S connectivity, the hardware landscape is where they necessarily fall behind. Demonstrating the availability of products in the market and showing flexibility in how they are implemented on a device level is an important step for them in creating a market for themselves.

An Open Hardware Ecosystem Must Develop


There is no sense in bashing the satellite IoT ecosystem for its excessive ties to connectivity hardware currently. This is a natural and required phase in the growth of the market, and third parties will involve themselves in the market as more clarity comes around standardization of connectivity approaches and around what architectures work best within different IoT markets. In the chipset and module market, more flexibility and availability will mean more business for the simple reason that customers like choice; choice allows them to select specific features suitable for their own applications, putting downward pressure on prices, and ensuring availability.

All three of these choices are particularly important for the satellite IoT market now. First, the IoT has shown itself to have many facets, each with different feature requirements; feature choice is, therefore, critical. Second, satellite operators want to differentiate themselves from incumbents like Inmarsat, Globalstar, or ORBCOMM partly through lower hardware and connectivity costs. Currently, the hardware component of a solution remains fairly expensive: US$49 for an Astonode S module, US$30 for a Kinéis KIM module, or US$50 for a Myriota module. Downward pressure on prices is, therefore, important. Third, availability of components is a key concern for many device manufacturers, with lead times for wireless Microcontroller Units (MCUs) often longer than 1 year. It is difficult even for large incumbent semiconductor manufacturers to ramp up production, and even more difficult for vendors placing small orders, so satellite network operators have to find ways of ensuring ready availability of their hardware for their customers to test and develop. One satellite company recently noted that limited component supply meant that it could only provide one or two testing kits to the many customers asking to trial its solution. The incoming interest is clear, but the inability to ride the wave or clear backlogs can lead to disillusionment with the technology or, at the very least, with the supplier. Finances for many satellite companies are already resting on perilous foundations thanks to the dry-up in technology investment and the crowded market, so operators have thin margins indeed with which to play.

An open hardware ecosystem is the clear way for the market to develop. The announcements above and the shift in messaging around satellite IoT may herald the first steps in this direction.