Could LTE-M Torpedo NB-IoT?

Like a North Korean missile, NB-IoT was launched to the accompaniment of much chest-thumping rhetoric only to hit problems soon after take-off. Several months since it emerged from the lab, the Internet of Things standard risks being knocked off course by interoperability problems and service providers' preference for the rival, unlicensed-spectrum technologies it was supposed to "crush." (See Ericsson, Huawei incompatibility threatens NB-IoT – sources and Vodafone to 'Crush' LoRa, Sigfox With NB-IoT.)

But could the biggest threat to NB-IoT come from another cellular technology? The LTE-M standard, which also formed a part of the 3GPP's Release 13 update last year, was never expected to be of much consequence outside North America, says Tom Rebbeck, a director at the Analysys Mason consulting and market-research group. Yet it has been quietly attracting interest in other parts of the world. A major commitment to the standard by an Asian heavyweight like China Mobile Ltd. (NYSE: CHL) could blow NB-IoT out of the market.

At least, says Rebbeck, it would "make the case harder for NB-IoT" as the globally preferred cellular standard for so-called "low-power, wide-area" (LPWA) connectivity. And if the short history of wireless communications tells us anything, it is surely that standards addressing the same opportunities rarely coexist in harmony. Battles ensue, and sooner or later the industry gravitates toward a single technology -- LTE rather than WiMax, for instance, or GSM over CDMA.

The momentum behind LTE-M is undeniable. As expected, both AT&T Inc. (NYSE: T) and Verizon Communications Inc. (NYSE: VZ) are backing it in the US. More surprising is the support that LTE-M has gained in Europe in the last year. Although NB-IoT is favored by industry giants including Germany's Deutsche Telekom AG (NYSE: DT) and UK-based Vodafone Group plc (NYSE: VOD), several big operators have swung behind LTE-M, including France's Orange (NYSE: FTE), Spain's Telefónica and KPN Telecom NV (NYSE: KPN) of the Netherlands. Both KPN and Orange are using LTE-M as a complement to LoRa, while neither has announced plans for NB-IoT. (See Verizon Takes IoT Network Nationwide, DT Claims World First on NB-IoT, Tier 1 Operators Back LTE-M, KPN to Include LTE-M in IoT Mix in 2017 and Eurobites: Orange Bangs the Drum for LTE-M.)

South Korea's SK Telecom (Nasdaq: SKM) is similarly following that approach of building an LPWA strategy around LoRa and LTE-M. LoRa, it previously told Light Reading, is ideal for "low-mobility" applications and transmitting small amounts of data over long distances. LTE-M, by contrast, is better for connected car services and others that are a bit hungrier for bandwidth. (See SK Telecom Sees LTE-M, LoRa as Its 'Two Main IoT Pillars'.)

If, from a pure technology perspective, NB-IoT's closest rivals are Sigfox and LoRa, then operators might be expected to use it alongside LTE-M in a similar, complementary fashion. That is certainly how the vendors are presenting it. "LTE-M is not a threat to NB-IoT -- it's actually complementary to NB-IoT," said a spokesperson for China's Huawei Technologies Co. Ltd. in comments emailed to Light Reading.

But Rebbeck disagrees. In his view, the growing appeal of LTE-M has led some operators to hold off making any commitments to NB-IoT. "The price difference between NB-IoT and LTE-M modules is marginal and there is not much difference in terms of battery performance," he tells Light Reading. "And the performance of LTE-M is a lot higher and might be more flexible for future applications."

Want to know more about the Internet of Things? Check out our dedicated IoT content channel here on Light Reading.

That sounds like a compelling reason for a service provider to consider using LTE-M instead of NB-IoT. And some operators evidently believe LTE-M can address opportunities often discussed in the context of NB-IoT, Sigfox and LoRa. Orange, for example, has been carrying out a pilot of smart electricity metering using LTE-M. Verizon says LTE-M can support "an array of use cases ranging from water meters to asset trackers to consumer electronics."

So perhaps, unlike SK Telecom, Orange has settled on LoRa and LTE-M not because they address radically different opportunities, but simply to make sure it has both licensed and unlicensed spectrum options in its LPWA arsenal. As a 3GPP-sanctioned standard, LTE-M runs over licensed spectrum bands, of course, while LoRa takes advantage of unlicensed frequencies. (Technically, it seems, there is nothing to prevent LoRa from being used in licensed spectrum bands, but existing equipment and regulations pose barriers to this move, Orange told Light Reading in December last year.) That would not appear to leave much room for NB-IoT. (See Is LoRa Going Licensed?)

NB-IoT is still very much alive and kicking, despite the challenges it faces. But the industry now looks at LTE-M far more approvingly than it did just a year ago, when, says Rebbeck, the technology seemed likely to be confined to North America. It has now gained supporters in Europe and the Far East, he says, while "the Chinese operators are playing around with [it]." A big commitment to LTE-M by the likes of China Mobile might spell further disappointment for NB-IoT.

— Iain Morris, Circle me on Google+ Follow me on TwitterVisit my LinkedIn profile, News Editor, Light Reading

ChiefTec48278 5/26/2017 | 12:17:38 PM
Re: LTE-M versus NB-IoT: existing v.s. new business model Hello Mr. Zune, you are clearly an expert in this field. I would like to reach out to you about a potential opportunity. Would you please email me at justin at my company's domain? Www.lynqme.com. I'm the CTO. Thanks!
Mr. Zune 5/10/2017 | 9:49:52 AM
LTE-M versus NB-IoT: existing v.s. new business model I have practical experience with both LTE-M, NB-IoT and LoRA.

Both 3GPP-technologies are very complementary to each other from a capability point of view, but for some use cases they can compete, but also combined!

Good to understand is that LTE-M is a service as we all know today with setting-up a wireless mobile IP connection, (like with 2G (GRPS), 3G (HSPA) and 4G. Off course with some nice features, like PSM and eDRX, eTAU. (same for NB-IoT) Also the service model is/can be simply volume (cost per MByte) based billing. Although it is expected that it will be more expensive per bit, than regular 4G data, because of more spectrum/network resources per transmitted/recieved bit.

NB-IoT is not really designed/meant for end2end (user) IP connectivity, the capacity is simple to small. But it offers a much richer, efficient way of communication with non-IP communication protocols. More IoT LPWA device protocols, like MQTT, COAP to NB-IoT network, not as an end2end service to the device over IP. It feels more like LoRA over a mobile network, without the hassle of message duplication and with guaranteed data delivery and much better coverage. (if deployed on existing mobile network)

This means NB-IoT is better for battery/power consumption and has much better coverage and higher reliability, but it has lower throughput. It is also not meant for continues (low latency) connectivity. (oppose to LTE-M or 4G)

Hence, much more designed for large scale deployments with irregular Tx/Rx and data aggregation can be very simple and cost-efficient (like a query in database). BUT this means, a complete differently business/operating model for the mobile operator, billing in volume (Mbytes) is useless, but a mobile operator can provide a complete working connectivity solution with or without added values, like network based geo-location or network based triggers. (like movement) or other IoT services. (on data/applicaiton e.g.) The extra positive thing for "the customer"is reduced hassle/worry on (IP) security wit the lack of a TCP/IP stack on NB-IoT device, so similar to LoRA, SigFox.

But this means that both NB-IoT billing/business model is completely different than today's business of a mobile operators, billing can be for example on the (limited/bulk) amount messages transferred (like with LoreWAN), or per active device, per use-case of final IoT product.

This means that mobile operators are likely to (internal) struggle to enter the NB-IoT market to create a real proposition/product, because of lack of (non-IP) LPWA IoT experience. They need adapting to this new (but attractive) model. Mobile operators can "taste this new model freely" with trying/deploying first using LoRa(WAN) technology and later on migrate to NB-IoT and/or try out first LTE-M, since LTE-M fits today's business model of mobile operators. But whether LTE-M really enables IoT on a large scale is more doubtful. LTE-M is (more cost-efficient) replacement for 2G GRPS for today's M2M business (nice when fasing-out 2G) where NB-IoT is a real enabler for massive (ubiquitous) IoT solutions.

CEO/Co-f00573 5/3/2017 | 1:04:27 PM
I don't see it, either 1. LTE-Cat M1 is not low power and we should not include it with other LPWAN forecasts. It's a replacement for GPRS and needs mains power. Maybe it can be called a MPWAN or HPWAN or something similarly unimaginative. http://bit.ly/2noxibV

2. NB-IoT is a non-starter for most enterprise/industrial use cases and forecasts should be tuned accordingly. http://bit.ly/2noxibV 

3. Maybe cellular LPWAN's are overhyped? Just thinking out loud here ...
DanJones 5/2/2017 | 3:53:03 PM
I don't see it. LTE-M requires a LOT more transmit power as it can support downloads at 1-Mbit/s. NB-IoT is scaled down to 100-200 kbit/s so it can support battery life in the years scale. NB-IoT is being overtaken by SigFox and Lora because they're available now, simple as that.
Sign In