3GPP—the force behind the standardization of cellular systems—recently introduced three new standards: LTE-M, NB-IoT, and EC-GSM-IoT.
The cellular industry's response to the LoRa, LoRaWAN, and Sigfox-type technologies, these new standards have been created to enable devices that operate on carrier networks less expensive and more power efficient. All three standards are slightly different from one another, yet each is fighting to become dominant.
Below, we’ll describe those differences and explain what you need to know about the technology and uses.
LTE-M (LTE-MTC, LTE Category M1)
LTE-M—an abbreviated version of LTE-MTC (or "machine-type communications")—is part of 3GPP’s release 12 and 13, finalized in 2016. It’s also referred to as LTE-MTC or LTE Cat M1.
In the simplest terms, LTE-M is a stripped-down version of LTE. It uses the same spectrum and basestations, works everywhere that LTE works, and enables true TCP/IP data sessions. The major difference between LTE and LTE-M is power efficiency—LTE-M enables battery-powered devices to send and receive data online via a Verizon or AT&T connection. An iPhone battery lasts a day, but a cell modem-connected water meter battery could last 10 years—which is a profound change to the cellular Internet of Things.
LTE-M has a slightly higher data rate than NB-IoT and EC-GSM-IoT (which we’ll discuss later in this article) but is able to transmit fairly large chunks of data. Potential applications include asset tracking, energy management, and utility metering; this technology could also be used in city infrastructure and wearable devices.
The energy efficiency of LTE-M is due in large part to its ability to remain virtually "attached" to the network while not physically maintaining a connection. There are two modes that enable this: power-saving mode (PSM) and extended discontinuous repetition cycle (eDRX). PSM allows LTE-M devices to go idle without having to rejoin the network when they wake up. eDRX allows endpoints to "tell" the cell tower or network how often it will be awake for downlink, which could be anywhere from 10 seconds to more than 40 minutes. For a deeper dive on these features, take a look at this recent webinar.
Advantage In The Marketplace
The advantage of LTE-M for M2M communications is that it works within the normal construct of LTE networks. In other words, a cellular carrier like Verizon or AT&T only has to upload new software onto its base stations to enable LTE-M, and won’t have to spend any money on new antennas. It’s also much simpler than a category 4 receiver (like that found in cell phones) because it only needs to understand and digitize 1.4 MHz of the channel instead of 20 MHz.
Suggested Reading: Link Labs offers the world’s first end-device-certified LTE Cat-M1 modem for battery-powered applications
NB-IoT (LTE Cat-NB1, LTE Cat-M2)
Narrowband IoT (NB-IoT) was added in 3GPP’s Rel 13. Unlike LTE-M, NB-IoT is not related to LTE, but is based on a DSSS modulation similar to the old Neul version of Weightless-W. Plenty of large telcos—including Huawei, Ericsson, Qualcomm, and Vodafone—are actively involved in putting this standard together, but it is not currently deployed anywhere in the U.S.
The fabrication of the chipsets in LTE-M and NB-IoT is nearly the same, so there’s not much that is fundamentally cheaper about using NB-IoT from an endpoint hardware perspective. The primary issue to consider with NB-IoT is deployability. Telcos believe that NB-IoT will help them compete at the low end of the IoT market against technologies like Sigfox and LoRa—but unlike LTE-M, there’s no path to a simple software upgrade to support this standard. Proponents of the standard will have to spend billions of dollars to support the technology—and it remains to be seen whether they’ll do so. It is most attractive option in places where existing LTE networks don’t already exist, as the basestation hardware for NB-IoT will likely be less expensive than LTE basestations.
Dish network, for example, has publicly announced plans to build NB-IoT networks. That said, Dish will likely never build a network the size of Verizon, so it remains to be seen how this technology will be used and adopted in the IoT community.
NB-IoT has similar power savings features to LTE-M (PSM and eDRX), but it lacks mobility at this time (which means it is primarily suited to static applications).
Suggested Reading: An Overview Of Narrowband IoT
EC-GSM-IoT is similar to LTE-M in that it is designed to operate in existing networks (in this case, eGPRS 2G networks rather than LTE 4G networks). It is also designed to operate with existing basestation hardware, to minimize the costs of upgrading eGPRS-only to support it. The details behind EC-GSM-IoT remain a little hazy, as it is not yet part of an official 3GPP release—but it is likely to feature eDRX, as well as some protocol-level enhancements (which may improve coverage by 10-20 dB). It isn’t yet clear there is market demand for this technology, and we don’t expect to see commercial EC-GSM deployments until 2019, if at all.
Comparison Of Cellular IoT Technologies
The question on everyone’s mind is this: Which cellular IoT standard will win?
The answer, at this point, depends on who you ask.
NB-IoT could potentially be less expensive (though not by much), but most carriers will not be able to use the same LTE radios they use for their data networks today. That said, LTE-M supports TCP/IP-based communications, while NB-IoT is message-based with a payload and response—which makes firmware upgrades much more difficult for NB-IoT.
Carriers that use Huawei basestations—or those that do not yet have large-scale LTE deployments—will be incentivized to support NB-IoT. (It’s worth noting that Huawei is building LTE-M modules, so they are hedging their bets.) Carriers that already have extensive LTE networks will likely take on LTE-M first. In many cases, operators will support both and see which standard sticks with customers.
It’s important to mention that LTE-M is the most mature in terms of deployment, networks, and in hardware. NB-IoT chipsets are still in early stages, with only a few telcos testing and piloting in Europe and Asia. If you’re two to three years out with your deployment, be sure to watch this space! If you need to go to market now, LTE-M could be a good option for you. Be sure to research all your options before selection—and get in touch if you’d like some assistance!