The phrase "alphabet soup" is regularly applied to the telecoms sector -- and with good reason. Now, developments in wireless connectivity options for machine-to-machine (M2M) communications and the Internet of Things (IoT) have unearthed a plethora of new acronyms that are certainly not for the faint hearted.
This has been particularly evident of late for IoT technologies dubbed low power wide area (LPWA) networks. These networks are deemed a critical element in the platforms that will in future support the burgeoning range of connected objects with long battery lives and low data rate requirements.
Steve Hilton, co-founder and managing director of research and analyst firm MachNation, describes it rather well: "The LPWA world is a bit like a nicely prepared minestrone," he said. "Stir the bowl a little bit and up pops another surprising vegetable you didn't see a moment ago."
For sure, competing technologies continue to emerge in the LPWAN segment.
Proprietary technologies such as Sigfox, Ingenu and LoRa have already made clear their ambitions to become globally available standards for public LPWA networks using unlicensed spectrum. They now face increasing competition from so-called cellular IoT technologies that fall under the 3GPP standardisation umbrella and will operate in licensed spectrum.
Aapo Markkanen, principal analyst at Machina Research, also notes there are more technologies still. "Weightless-N, Weightless-P, NB-Fi (WAVIoT), Accellus, Flexnet (Sensus), Telensa UNB, and Synergize (Aclara) come to mind," he said. "And those are only the ones that are LPWA in the real sense of the term. You can also find a bunch of others that are going after many of the key applications, but they've built on a mesh architecture so they sit under a different technology umbrella."
Cellular IoT developments have also contributed significantly to the "minestrone" of standards and acronyms in the LPWA field. Here, David Hammarwall, Ericsson's head of 4G/5G services and infrastructure within the Swedish vendor's radio business unit, summarises the current situation by saying: "EC-GSM, NB-IoT [now renamed LTE Cat-M2] and [LTE] Cat-M1 are the main [cellular IoT] LPWA contenders. It depends on the operator's use cases and network evolution strategy which is the best option for them."
Proprietary vs. standards
The question is: How many of these various standards will exist in future? Will proprietary options eventually be pushed out by the cellular standards, or can a number of different standards continue to co-exist in future?
Narrowband IoT (NB-IoT) has certainly been causing a stir among the traditional mobile vendor and operator community. Standards are not yet in place, but 3GPP work on the Cellular Internet of Things (CIoT) is due to be completed before June to enable its inclusion in Release 13.
Arne Schaelicke, who leads LTE marketing for Nokia's Mobile Networks, explains why standards such as NB-IoT are so compelling for the traditional mobile players. "Sigfox, LoRa and Ingenu use unlicensed spectrum. As everybody could use this spectrum, there could emerge even more competing technologies. The more technologies and the more devices connect on this spectrum, the higher the risk of interference. Only cellular IoT technologies on licensed spectrum will allow for reliable IoT connectivity in large areas in the long term."
MachNation's Hilton agrees. "Carriers like to build standards-based networks for the same reasons that good restaurants follow recipes -- having predictable outcomes and minimising risks is really important when you are building for the long-term mass market."
Markkanen of Machina Research added that NB-IoT "is what the majority of operators are prioritising at the moment -- and to my knowledge most are planning to tuck it in the guard bands of their LTE networks. GSM isn't really relevant in this context anymore."
At the same time, Peter Jarich, vice president for consumer and infrastructure services at Current Analysis, noted that despite the growing interest in Cat-M2, there is no technical reason why the different technologies, proprietary and otherwise, could not live alongside each other, although not all of them are likely to survive.
Indeed, even as operators declare their support for and interest in NB-IoT/LTE Cat-M2, they are investing in non-cellular alternatives in the meantime.
As explained by Yves Bellego, director of technical and network strategy at Orange: "We deployed LoRa technology in France because it provides bi-directional connectivity. One benefit of these technologies is to be available now for commercial service."
However, Bellego added: "We expect that cellular-based solutions will become the most used in few years, but proprietary technologies that are being deployed today will have a good lifetime."
While this indicates that mobile operator interest in proprietary technologies will be short term, Jarich pointed out that service providers "aren't the only players in town."
"In other words, IoT connectivity won't necessarily be a service provider-only thing; it's not up to service providers alone as to whether or not IoT deployments leverage cellular or non-cellular technologies," he said.
Business considerations, timing play roles
Ultimately, technology and business considerations will play a role in deciding which technologies will be favoured over the longer term.
According to Jarich, from a business perspective "we need to acknowledge the different models at play here: Sigfox and Ingenu are building out networks, LoRa is building an ecosystem. That means that each will appeal to different use cases. On the technology side of things, you've got Ingenu claiming that RPMA is better suited to robust IoT services. Again, however, it's a question of 'horses for courses' when you consider the needs of different use cases -- not every IoT sensor needs a capable downlink, for example."
Hilton pointed out that timing will also play an important role in determining the longevity of the different technologies, particularly the proprietary solutions: "If actual 3GPP solutions take a long time to materialise, then one or more of these standards might become the next iDEN. It will grow and become relatively dominant from a technology and a customer perspective," he said. "On the other hand, if 3GPP solutions fairly quickly materialise, then all these proprietary solutions…will end not with a bang, but a whimper."
Ultimately, LPWA is only one element of a much bigger connectivity picture that will be dominated by a more heterogeneous network structure than in the past -- particularly as the industry moves toward 5G.
According to Markkanen, "it is such a vast and diverse space that there simply is no way that it would see a similar consolidation to what took place in mobile communications, where technical requirements were greatly more homogenous to begin with. The solution to managing all that 'fragmentation' -- also known as 'choice' -- and problems that it causes is purposely designed platforms and other middleware."
For enterprises making buying decisions, added Hilton, "the bigger issues are those surrounding the upper layers of the IoT technology stack like IoT application enablement platforms, platform-enabled solutions, technology integrations, IoT cyber security and IoT orchestration."
These are much bigger challenges for the companies that buy IoT solutions than which LPWAN they should use, he concluded.