It’s time for chips that combine cellular and unlicensed bands

Companies making low-cost sensors that rely on battery power have long turned to a handful of wireless technologies available in what’s known as the unlicensed spectrum bands. Bluetooth Low Energy, ZigBee, and Z-Wave each offer a way to connect devices in resilient and low-power fashion, but can’t connect back to the internet. For that, you need a hub in the form of a gateway device or a cellphone.

For devices that don’t stay put, or are reliant on a dedicated and more secure cloud connection, manufacturers have had to turn to cellular technologies or dedicated Low-Power Wide-Area Networks (LPWAN) such as Sigfox. The challenge with these networks is that the radios sometimes cost a bit more and may come with subscription fees that other options don’t have. For most devices, you’ve had to pick a licensed cellular technology or an unlicensed one.

But as more devices come online and demand for wireless connectivity that can reach anywhere expands, some are seeing demand for a combined Bluetooth and NB-IoT radio. Or maybe a combined Sigfox and Bluetooth radio. For example, a shipping container might use a cellular connection at sea or on the road but transition back to Bluetooth when in a port or a storage facility.

The transitions between radios could save on data costs, and maybe energy use, too. Inside an office, a Bluetooth radio might track equipment, while a cellular option could make sure that equipment doesn’t get lost when an employee “borrows” it or even moves it between buildings.

Nordic Semiconductor already offers a combined Bluetooth and cellular functionality in its dev kits, including with a few reference designs that use a Bluetooth chip and a cellular module. But to get the most benefits and the greatest cost reductions, the goal would be to integrate both radios on the same chip. Doing so would cut power usage and cost, and would allow manufacturers to shrink the end product. But such integration doesn’t happen on a whim — chipmakers have to be sure demand is there before undertaking the challenge.

When I sent an email to Svein-Egil Nielsen, CTO of Nordic Semiconductor, asking about new, integrated radios, he sent back the following:

“[I]f the history of technology teaches us anything it’s that the trend is always towards ever-increasing integration. So, will there one day be a market-driven demand for a combined Bluetooth and cellular (NB-IoT or LTE-M or both) single packaged solution? I would say (very) probably.”

Nor did he rule out the possibility of a combined licensed and unlicensed IoT wireless solution that combines cellular and LoRa/Sigfox. But he’s also unsure if technologies such as LoRa will survive the rollout of full-scale cellular LPWA networks.

On the engineering side, the combination of the radios is feasible; the biggest challenge would be figuring out how to integrate the different antennas required for each radio. There’s also a difference between NB-IoT and the LTE Cat-M standard, which has a larger data rate but is still power efficient. NB-IoT radios are best suited for devices that don’t move, such as smoke detectors, parking spot sensors, and road sensors. Whereas LTE Cat-M can handle mobility, which makes it more appropriate for something like the shipping container example mentioned above.

As the list of device types we want to connect continues to expand, our technologies for connecting all those different device types will expand as well. After all, a decade or so ago it was crazy to imagine a cellular carrier getting excited about Wi-Fi offload, but now it’s automatic on most handsets. Cellular and unlicensed bands already co-exist. The IoT will just deepen that integration and incorporate more radios.