The U-LPWAN border conflicts continue this week, with the factions announcing projects, promoting mouthpieces, and claiming new capabilities. Incapable of uniting against their shared enemy L-LPWAN, the U-LPWAN sector is still slogging it out, trying to claim points and wins instead of putting on a collective front.
Of course, there are very strong business motivations preventing them uniting, but unlike fights like VHS vs. Betamax, there’s still plenty of life in the market for these technologies if they don’t take first place. With an immense number of pivot opportunities, it doesn’t look like there is much in the pipeline that could kill one of these U-LPWAN technologies dead, besides a company selling up, investors pulling out, or perhaps a political scandal. Ingenu’s continued existence might be the best proof of this.
From the recent spate of news, perhaps most pressing is an announcement from BehrTech (Behr Technologies, also BTI), where an independent study has found that its MyThings U-LPWAN technology (formerly MIOTY) outperforms LoRaWAN. We covered the introduction of BehrTech back in August 2018, and while it is a mesh network approach, more like Wi-SUN than the star topologies used by LoRaWAN and Sigfox, it is competing in the same industrial markets that so many LoRaWAN advocates are pursuing.
To this end, BehrTech wants to prove that its technology has superior QoS to LoRaWAN, its chief rival, and so it funded an investigation by a professor from the Nuremberg University of Technology for Applied Science. This found that in “a real-world Industrial IoT environment with high levels of interference from other systems and devices, MyThings has considerably higher interference resilience than LoRa.”
The bottom line is that the test found that LoRa had a 5% Packet Error Rate (PER) in the ‘real-world’ testing environment, and in the ‘dense interference scenario’ this reached 10%. In both, MyThings apparently lost zero messages. When there is no interference, the test found that network performance was nearly the same, however.
As such, BehrTech concludes that if you have a mission-critical application, you can’t rely on LoRaWAN. Of course, there are plenty of other applications that don’t have that absolute requirement, and plenty of ways around the problem of a failed message send, but once you begin evaluating the cost of failure in a mission-critical system, you do begin to gravitate towards hard-wiring everything in to avoid problems with radio waves traveling through the air, but that’s a very slippery slope.
The interference was simulated using the IEEE’s LPWAN Interference Model Standard, and the test had four different levels for interference – None (a=0), Low (a=1, aking to a smart building), Medium (a=10, akin to a smart city), and Dense (a=50, the Industrial IoT scenario).
For some context, the Dense environment is based on the assumptions that there are 7,000 households per square-kilometer, with 100 connected devices each, therefore giving 700,000 devices per square-kilometer, with the devices sending a 5-millisecond message every 10 seconds. For the LoRaWAN connection, a Multitech gateway operating on the SF12 spreading factor was used, and the MyThings Pilot Kit was used for BehrTech’s option. Both were running in European modes.
The study was announced in the same week that LoRaWAN vendor Senet proclaimed that LoRaWAN was the “global standard for Industrial IoT,” following a deal with systems integrator SimplyCity to deploy LoRaWAN networks for city councils in Australia. SimplyCity is also deploying an environmental monitoring system in a salt mine in Western Australia, prompting Senet to declare that “this flagship project will set the global standard for industrial-grade IoT projects at remote locations under harsh conditions.”
So, good luck discerning some objective truth from the two camps. Anecdotes, personal experience, and politics to some degree, will all influence any given person’s opinion on the best LPWAN option for the job. But one thing that seems to have solidified in the past year or so is that Sigfox is playing at a different end of the scale to the rest of the crowd.
Sigfox has updated the world on its current numbers, by way of getting a new PR company. It says it has coverage in 65 countries, covering 5mn square-kilometers and 1bn people, and totaling 10.3mn devices that send 20mn daily messages. We are also aware that the Eutelsat nanosatellite project is progressing, albeit a bit slower than anticipated.
In terms of progress, Sigfox said in November it had 53 countries with live networks, but it was also saying 1bn people covered back then. Adding 12 new countries should have increased that population coverage somewhat. Back in February 2018, when Sigfox announced the major climb-down, it had covered 45 countries with 800mn people under the umbrella.
The daily messages rate is also notable, with the average Sigfox device sending just under 2 messages per day. There are too many variables in play to begin calculating power usage and battery life, and the same can be said for the number of messages actually sent daily – as many Sigfox devices are used in out-of-band (OOB) applications where dormant operation is routine until something unexpected happens and a flurry of messages are sent.
Sigfox has had a fairly busy summer, announcing a manufacturing partnership with Alps Alpine that will add Sigfox connectivity to many components and devices, as well as the re-signing of the Securitas Direct deal for 15-years in Spain and Portugal. Kryton just announced that it would be using Sigfox in its real-time concrete monitoring products, and Sigfox’s Hacking House keeps churning out startups with ideas that just might eventually drive volume deals for the technology.