MIOTY Silicon Vendor Agnostic, Scalable LPWAN Standard to Take on LoRaWAN, NB-IoT

[April 6th update: The article was updated to reflect Fraunhofer invented the technology]

There are plenty of LPWAN (Low Power Wide Area Networks) standards designed for low power, low bitrate and long-range connectivity with the most popular currently being NB-IoT and LoRaWAN. But Texas Instruments has joined other smaller companies (Fraunhofer, Ragsol, STACKFORCE, WIKA…) to form the MIOTY alliance in order to develop and promote a new LPWAN standard operating in the sub-GHz range called MIOTY.

MIOTY

Texas Instruments explains MIOTY was born due to a lack of scalability and robustness (interferences issues) of current LPWAN solutions. But just like other similar low power WAN standards, it aims to support long-range while achieving long battery life, but MIOTY is also supposed to help IoT developers mitigate performance degradation in high-node-count networks.

MIOTY works in the same license-free bands (868 MHz, 915 MHz …) as LoRa radios with no costs involved to use the radio spectrum. The solution offers a star network for low-power end/leaf nodes, as well as a gateway solution for cloud connectivity. Currently, companies have to set up their own private networks, but eventually, there may be third parties offering network solutions.

LoRaWAN only works on chips developed Semtech, while Sigfox also relies on proprietary solutions, but MIOTY physical layer (PHY) and media access control (MAC), but MIOTY is based on a standard from the European Telecommunications Standards Institute (ETSI), namely TS 103 357 “Short Range Devices; Low Throughput Networks (LTN); Protocols for radio interface A “, which can be downloaded free of charge.

Texas Instruments CC1310 Cortex-M3 microcontroller, released in 2015 with the promise to offer up to 20 km range, 20-year battery life on a coin cell, just happens to be compatible with MIOTY standard, although the company also mentions MIOTY was tested on two other silicon providers. At the time of writing, I was unable to find out who the other two are. Other compatible chips from TI include CC1312R and the CC1352R Cortex-M4F WiSoCs with the latter also supporting Bluetooth Low Energy.

Back in 2018, Canada based BehrTech partnered with Fraunhofer IIS to commercialize the solution, and the key ingredient for robustness (resistance to interference) is Fraunhofer IIS’ patented Telegram Splitting technology that split messages into several packets to increase immunity to interference. A whitepaper quickly explains the history of MIOTY standard:

MIOTY by Fraunhofer Institute for Integrated Circuits (IIS) was adopted by European Telecommunications Standards Institute (ETSI) as Low Throughput Network (LTN) family Telegram-Splitting Ultra Narrow Band (TS-UNB) [5], which is based on Gaussian Minimum-Shift Keying (GMSK), a variant of FSK. The nominal range is 15 km and the battery lifetime is stated as up to 20 years. So far only one company, Behr Technologies, is commercially marketing MIOTY technology; on the gateway side they use an SDRplay RSP1a  Software-Defined Radio (SDR) chipset with proprietary userspace libraries.

The 2018 video below quickly explains how that all works.

One way to get started with the new standard is to get TI LPSTK SensorTag kit.

Support CNX Software - Donate via PayPal or become a Patron on Patreon
Advertisements

6
Leave a Reply

avatar
5 Comment threads
1 Thread replies
0 Followers
 
Most reacted comment
Hottest comment thread
4 Comment authors
DroneDavid WillmorezoobabPhilipp Blum Recent comment authors
  Subscribe  
newest oldest most voted
Notify of
Philipp Blum
Guest
Philipp Blum

Okay, let’s see where this is going. Could be potentially be more interesting than lora. But there is a long way to really cheap modules and a good infrastructure. They also need something like TTN.

zoobab
Guest

“Fraunhofer IIS’ patented Telegram Splitting technology that split messages into several packets to increase immunity to interference”

Patents are the nail in the cofin.

Including for Lora.

We suffered 20 years of Fraunhofer MP3 patents.

David Willmore
Guest
David Willmore

Agreed. The only benefit they provided was pushing the need for license free codecs.

“It could be that your purpose in life is to be a warning to others.”

Drone
Guest
Drone

“The solution offers a star network for low-power end/leaf nodes…”

What’s really needed is support for mesh (not star) topology, or at-least a hop-on repeat function. Without this, building out infrastructure is expensive.

“…they use an SDRplay RSP1a Software-Defined Radio (SDR) chipset with proprietary userspace libraries.”

I think the SDR1a uses off the shelf chips. I’m working frpm vague memory bit I think the IQ RF chip is made bt Mirics, the ADC is made by ADI, and the USB interface can be made by anybody.

Finally, Telegram Splitting is encumbered by a Fraunhofer patent.

Drone
Guest
Drone

Here’s a MIOTY intro YouTube video (in German) from Fraunhofer IIS that’s dated Mar 10, 2016. So yeah – while the MIOTY Alliance might be new, the technology has been around for awhile.

https://www.youtube.com/watch?v=f2mL-J1zFJs

Drone
Guest
Drone

BEHRTECH of Toronto makes a LPWAN system called MYTHINGS that looks the same as MYIOT (both are based on Fraunhofer’s Telegram Splitting – Ultra Narrow Band [TS-UNB], ETSI TS 103-357).

BEHRTECH puts out a report titled “MYTHINGS vs. LoRa, A Comparative Study of Quality-of-Service Under External Interference, Conducted by Professor Dr. Thomas Lauterbach.” Supposedly you can download the report here:

behrtech.com/resources/lora-vs-mythings/

Unfortunately I could not get the BEHRTECH form you must fill out before you can download the report to work. Also, the form demands FAR too much personal information! (Talk about sales prevention.) So I did a web search looking for an end-run around, and found one. You can download the BEHRTECH report here:

https://iotbusinessnews.com/download/white-papers/BEHRTECH-MYTHINGS-vs-LoRa.pdf

The report says:

“This study will prove that in real-world Industrial IoT deployments (IIoT), MYTHINGS delivers
significantly higher interference resilience than LoRa. More specifically, it will show that in a high
interference environment at the same network range, MYTHINGS does not lose a single
message, while LoRa loses more than 10 percent of total messages and that when high
interference exists, there is always a certain extent of failed LoRa transmissions, regardless of
signal power and range.”

The report goes on to describe the test setup, but with very few details about the type of interfering signals and/or noise added. Without this information, the tests are impossible to reproduce or interpret; the first of several fatal flaws. Next, the report presents the results.

The results show that with no interference and at low packet error rates (the area of most interest) LoRa performs essentially the same as MYTHINGS. With high interference and at low packet error rates (PER) MYTHINGS outperforms LoRa by around 10dB in relative signal levels. However, when the MYTHINGS and LoRa test results are plotted on the same graph with the same scales, even at fairly high signal levels the LoRa PER never goes below around 3% (strange) while the MYTHINGS PER drops sharply to unmeasurable levels (as expected). To me this disparate behavior screams out: “Something is wrong with either the particular LoRa system used (MultiConnect Conduit IoT Starter Kit by Multitech) and/or there’s a problem with test setup!

After seeing the test results, If I were the principle investigator I would NOT have published the report without resolving the unexpected and obviously flawed results from the LoRa system. I would also have tested MYTHINGS against more than one LoRa manufacturer, to not do so leaves the door open to “cherry-picking” your opponent.

My conclusion is that this report comparing MYTHINGS vs. LoRa is not to be trusted. The only conclusion I took away from the report is that without any interferers present, the performance of MYTHINGS and one particular manufacturer’s LoRa system seems to be similar, but that’s it.

Maybe this explains why BEHRTECH makes it so difficult and intrusive download this report – they really don’t want anyone reading it.

Advertisements