LPWANs are dead. Long live LPWANs!
I was recently asked to comment on the Nokia’s whitepaper on LTE-M (Nokia proposal LTE-M) and found myself pondering two questions. Why has the cellular industry been late to the IoT game? Are LPWANs using unlicensed radio spectrum (termed here as LPWAN-U) no longer needed?
Regarding the last question, I came to the clear answer of “No, they are needed” and in fact LPWAN-U is yet to have its Heyday!
The cellular industry has been brilliant at provisioning connectivity for smart phones and enablement of mobile applications. However, when it has come to IoT, acting to “bridge” the IoT sensor “connectivity gap” to help facilitate the IoT to its full potential, they have been sluggish. Very sluggish.
Having tried for years to persuade industry and government that 3G and 4G services were ideally suited for the IoT, the cellular industry has changed tack. Enormously. Thus, we are starting to see proposals from the cellular industry for NB-LTE-M and LTE-M – termed here as LPWAN-LS – Low power wide area networks using licensed radio spectrum.
The LPWAN-U business communities have been speaking for years (Cambridge - 2010, France - 2010, US – 2012) on the need to bridge the “connectivity gap” with a new “IoT-oriented” long range, low power, low cost wireless technology paradigm. Call it “cellular IoT" if you like (I was using this term years ago). The IoT “connectivity gap” had to be bridged if global markets had the opportunity to develop to their full potential using sensors that were low cost and low energy and could be deployed in their zillions over long distances.
LPWAN-U is probably the most disruptive innovation that the wireless industry has seen for years. Virtually free “cellular-IoT” that can be deployed at low cost that is independent of cellular. Wow. It could be viewed as a democratisation of IoT telecommunications if you like. No wonder Gartner estimate 3.1 billion LPWAN-U sensor/device connections by 2023, three times more than cellular.
So it came as no shock that the cellular industry has agreed with us, woken up and is trying to bridge the “connectivity gap” with some new 3GPP initiatives levering spectrum for which it has paid huge sums of money.
So, what might be the impact of this kind of proposal from a cellular tech giant have on the LPWAN-U industry? Probably nothing of notable business impact for at least a few years. That’s my opinion.
The LPWAN-LS standards are scheduled to be agreed by 3GPP Q2 2016. Lets face it their contributing organisations will disagree on various aspects so I anticipate a realistic first stable version to be released Q4 2016. Network software has to be developed and ICs developed. Let’s be generous and assume there are alpha versions ready end of Q2 2017. Trials have to be run, lets say 6 months, allowing for festive celebrations, lets say trials are completed and analysed in less than 1 year – say end Q1 2018. We are now more than two years out from now. There have to be modifications of silicon and network software in order to obtain a stable version for deployment. Add on another Q, puts us at Q2 2018. Now for roll out. Depending on size of country, lets roll out goes fast with no glitches and takes 9 months. This now makes LPWAN-LS available for commercial use Q2 2019. That is more than 3 years out from now.
So, probably, the LPWAN-U community has three free blissful years (save maybe some disruptive marketing from the cellular industry in 2018/2019) to continue to lead the world on our ability to develop technologies, products and solutions that are of enormous value for multitudes of potential customers.
So, if you are building IoT solutions comprising sensors connected over a long distance, do you wait 3 years for LPWAN-LS or start building economic value today with LPWAN-U? I humbly suggest the answer is fairly obvious and don’t forget the many other disruptive benefits of levering LPWAN-U.
Create your Heyday!
Matthew has been internationally recognised for his work pioneering IoT, LPWANs and inspiring the world “to imagine”
i think the decisive factor here is sensor's battery endurance and cost of sensor. having a licensed band network does have the stability advantage. but if LoRA can do it on the unlicensed band with 2+ years advantage due on sensors due to early european deployment. The game could be different this time.
Agree with you Matthew. Each company has its own strength and contribute the growth of IoT industry. After 3 years, LPWAN-U should be much more mature to face the new competition. The question will come: do we need a standard to combine all the technology like the CDMA, GSM world in the past?
@Graham Gilbert Ed Hightower, Thanks for the comments guys. I regularly undertake detailed analysis of the LPWAN competitive landscape and refresh conclusions as the State of Health of each major player as well as identifying new innovation opportunities. The LPWAN-U community has done a tremendous job and worked hard at bringing a variety of products and business models into the IoT Markets. So, I don't want to be positive or negative about anyone. What I will say is that I suspect that each LPWAN-U provider will find its position within IoT markets and territories. I don't think we will see one clear winner in the near term and that is a good thing. My deep hope is that we dont see a Betamax/VHS scenario as differing LPWAN-Us can co-exist serving different IoT market and territory needs. Sigfox and Ingenu have been great at raising money which has put some faith into the investor markets and exposed the value of LPWANs. LoRaWAN has been great at developing international partner eco-systems. Weightless has led the way in developing open IoT LPWAN-U wireless standards. All these LPWAN-U organisations are doing a super job at developing the LPWAN-U markets and enabling businesses to develop value with a "IoT cellular", a new wireless/energy/cost/economic development paradigm. Without LPWAN-U the IoT markets would stifle to grow. Do you guys have an opinion on which LPWAN-U provider is leading the way?
What's your opinion on Sigfox Matthew ?
Very well said Matthew. Which LPWAN-U approaches should end-user companies be following? Which ones are most likely to succeed?