Your browser is not supported. Please update it.

20 July 2020

Mioty plays resilience card in LPWAN game

Distant observers of the burgeoning IoT market and the associated LPWAN (low power WAN) protocols – for longer distance wireless communication in sub-GHz spectrum –  might be mystified to observe another protocol clambering aboard with some serious backing. This is, after all, a congested field where heated battles are being fought already between five contenders for low power communications, across campuses, cities and industrial parks in particular.

The new challenger is called MIOTY (a portmanteau of MY IOT) and is playing the robustness card against its rivals. Certainly, on cursory analysis, it looks better placed for providing near-guaranteed delivery of data packets under poor signal conditions with interference from obstacles or other radio communications in dense RF environments. It is also playing on scalability by transferring complexity to single central points, typically some base stations, where hardware can be expanded, while minimizing processing in the devices themselves.

Naturally, all the LPWAN standards have been designed to minimize computational burden in order to enable small form factors and batteries to last effectively for their full shelf life, up to 20 years. But having come along later, MIOTY has been optimized totally for low bit-rate communication across unreliable wireless networks, where error-free delivery matters more than latency or bit-rate.

Its key feature has echoes of adaptive bit-rate (ABR) streaming for online video delivery, where the data is split into chunks to optimize quality and bandwidth consumption across unmanaged IP packet networks, primarily the Internet. In the case of MIOTY, packets are split into smaller sub-packets at the sensor level, a relatively simple operation easily executed by a small processor. These sub-packets are then transmitted at different frequencies and times as available, with some duplication for redundancy, arriving randomly and out of order from multiple clients at the central station. This station employs a sophisticated algorithm consuming considerable hardware resources scanning the spectrum for all MIOTY subpackets and re-assembling them into the complete messages.

This is based on an underlying ETSI specification called TS 103 357, or more informally telegram splitting, operating in the 868 MHz band, which is licence-free in Europe, requiring just 200 KHz of bandwidth. The method was developed by Germany’s Fraunhofer Institute and licensed to Canadian firm BehrTech for global commercialization.

The duplication allows full recovery of the received data in the event of up to 50% of sub-packets being lost in transmission, far higher in practice than commonly used forward error correction (FEC) or packet re-transmission mechanisms.

This is certainly an improvement over the rival LPWAN mechanisms, but as we know from the history of communications, being the best technology is no guarantee of success, especially when others have already gained wide traction.

We have covered the others before, but briefly there are five primary contenders: LoRaWAN, Sigfox, LTE-M, NB-IoT and Wi-SUN (wireless smart ubiquitous network).

Wi-SUN, based on the IEEE 802.15.4g standard, is designed for frequent low latency communication lasting up to 12 seconds, with a focus on minimizing power drain in the client while resting. It offers data rates up to 300Kbps, which the Wi-SUN Alliance claims enables the lowest latency of all the options, down to 0.02 seconds. But none of these would be chosen for use cases requiring ultra-low latency such as autonomous driving, where 5G-IoT comes in.

LoRAWAN, typically configured in a double-star topology, with multiple gateways relaying messages between larger numbers of client devices and a single central server, is designed for more infrequent communication of potentially longer duration up to 128 seconds, again minimizing power drain during both resting and listening states. This has longer latency, which reduces ability for devices to receive on-demand commands but increases resilience.

NB-IoT is also optimized for infrequent communications of even longer duration, up to 600 seconds or even more. It is a 3GPP standard enabling operation either within licensed spectrum or unused resource blocks within a carrier’s guard band. This confers some benefits such as better security and scalability than LoRaWAN and Sigfox, also with superior indoor penetration, but with low data rates, best for stationary devices requiring high efficiency within licensed spectrum.

MIOTY was motivated partly by the finding that all these other LPWAN protocols could be derailed by obstacles in the communication path that can only be overcome, especially in dense urban environments, through over-provisioning of gateways to avoid the resulting ‘shadows’. It was announced in 2018 but has taken until now to gain momentum and attract a critical mass of research organizations and industrial firms, around the MIOTY Alliance, to make the case against the other LPWAN technologies.

The alliance was formed by the Fraunhofer Institute for Integrated Circuits (IIS), joined since by Texas Instruments, German industrial technology firms Diehl and ifm, along with Austrian oil-and-gas technology provider Ragsol, German embedded connectivity provider Stackforce, and the UK’s industrial sensor maker WIKA.

The main target field is industrial automation, aiming to introduce wireless communications into hostile environments such as mines and refineries where there is a growing need for mobility and wire-free processes. Apart from robustness, scalability is the main focus, with a single MIOTY gateway based on commodity hardware able to handle 100,000 sensor nodes and over 1,000 messages per minute, at least in a test.

While the Industrial IoT is a primary focus, it will be among utilities where MIOTY is almost guaranteed traction as the successor to the Wireless Meter Bus (M-Bus) standard that has underpinned smart metering in Europe, offering better coverage, scalability in dense environments and robustness.

The question is first how quickly it can collar the emerging market for non-time critical IIoT applications and then expand from that base in private networks to the wider IoT domain where those other protocols have already gained traction. For that it will need to attract further system integration partners and it will find the going much harder outside those two core areas of utilities and IIoT.