FEATURES - Strategic Solutions Series
November 25, 2015

LTE vs. LPWA: Why LTE is the Clear Winner in Terms of Cost, Quality, and Scale

If you follow the Internet of Things space, you may have noticed all the talk lately about low-power wide area networks. The argument from the LPWA crowd suggests that LTE is more than IoT needs, in terms of bandwidth and cost. Instead, LPWA proponents say, the world needs networks that are purpose-built for the IoT.

While that may seem a logical assumption at first glance, the fact is that cellular service providers and their suppliers already have technology to support lower cost, lower bandwidth, and more power-efficient LTE for the IoT – and that technology will work today on the cellular networks that are already widely deployed. Meanwhile, the LPWA contingent will have to build and finance their networks from scratch.

Another important distinction between LTE and LPWA is that while the former is based on industry standards and supported by a broad ecosystem, LPWA in its various flavors is based on an array of proprietary technologies – like Ingenu’s RPMA, the LoRa Alliance’s LoRaWAN, SIGFOX and its UNB technology, and probably another 10 companies and technologies that are trying to gain traction. That means those who commit to LPWA technologies will not benefit from the economies of scale and flexibility that a broad ecosystem like the one found with LTE can provide, points out Eran Eshed, co-founder and vice president of marketing and business development at Altair Semiconductor.

There is admittedly a relatively high cost associated with bringing standards-based solutions to market, because of the need to certify and demonstrate interoperability, he adds. The good news for LTE is that much of that work has already been done.

“When you turn on LTE, it always works,” notes Eshed. “And when you buy and turn on a certified cellular device, it always works.”

It remains to be seen how reliable the LPWA networks, which operate on unlicensed spectrum, will be, Eshed indicates. While licensed spectrum was expensive for the cellular service providers to acquire so they could build their LTE networks, the spectrum is clean, supports high availability and quality of service, and is already available (for voice and data services) and paid for by the carriers. Unlicensed spectrum, meanwhile, is very prone to interference, he says.

You may have heard some of the LPWA players promoting the fact that they have $2 or $3 chipsets, and suggesting that as a result their promised solutions are lower cost alternatives to LTE, Eshed adds. But the total cost of building a wireless network involves a lot more than just the chipsets. It also involves the expensive and frequently lengthy processes of acquiring sites, building towers or leasing space on existing towers, investing in core networks, arranging for backhaul, investing in management and billing systems or striking deals to outsource these capabilities, and more.

“LTE networks are there, and they generate good revenue and profits for the carriers,” Eshed says. “LPWA involves a purpose-built network, and you have to build it from the ground up.”

There’s one company in the LPWA arena, says Eshed, that has what looks to be a disruptive business model. SIGFOX says it expects to charge $1 per node per year. However, as Eshed points out, if SIGFOX signed up 20 million customers, that wouldn’t be enough revenue to turn up even one tower.

That said, LTE – and its Cat-1, Cat-0, and Cat-M variants – is the clear connectivity solution for North American IoT deployments that require affordability, security, scale, and faster time to market, says Eshed. Both Cat-1 and Cat-0 have already been defined by the 3GPP. Altair announced the availability of its Cat-1 solution in February, and it is expected to go live in the U.S. in the fourth quarter of next year. Cat-0, which offers even lower speeds and longer endpoint battery life, should be in networks starting toward the middle of 2016. In fact, Verizon Wireless announced such plans with Altair in September. And Cat-M should be in widespread use by 2017. (To be clear, Cat-M is different from LTE-M; the latter is a term being used by Huawei, but LTE-M is not a 3GPP-sanctioned technology.)

So while we may see LPWA deployed in pockets here and there for specific use cases, like to provide connectivity for a smart lighting installation, such networks are not positioned for nationwide deployment in the U.S., or elsewhere such as covering the U.K. or France, says Eshed.

“The LPWA guys do have a chance,” he says, “but it’s nowhere near the scale that LTE can deliver.” 

Edited by Ken Briodagh

Back to Homepage
Comments powered by Disqus