An interview with Arasch Honarbacht, Managing Director of Ubisys Technologies GmbH. The German smart home company is one of the most active supporters and co-developers of the Zigbee standard and wants to continue this commitment in the matter age. Honarbacht explains why.
Dies ist die Übersetzung eines deutschen Interviews. Zum Original bitte hier entlang.
Ubisys as a company is more familiar to industry insiders. Why is the brand still little known to the public?
Arasch Honarbacht: That’s because we are very much on the technical side. In many places, we are not even noticed because we license our technology to large companies, for example, which then market it under their own name.
What is this technology?
Honarbacht: We have been instrumental in the development of Zigbee over the last ten years and were also awarded “Outstanding Contributor” for this in 2018. Likewise, we are driving exciting features coming soon to market, such as Zigbee Direct and the next version of the R23 Zigbee stack. You may have seen our logo in the Apple WWDC 2022 keynote: We will be one of the first companies to offer a matter/Zigbee bridge – at the official launch of the standard.
“We will be one of the first companies to offer a matter/Zigbee bridge – at the official launch of the standard.”
Will this be your existing G1 gateway or a new product?
Honarbacht: There will be a firmware update for the gateway G1. After that, our customers will have full matter functionality. We already have an Alexa Skill and support for Apple HomeKit, but I think matter will replace these individual integrations in the long term. At most, they will still be needed to keep existing installations running. We also license the bridge to other manufacturers who integrate it into their hubs.
Why Zigbee? With version 3.0 of the standard, compatibility has improved, but there are still manufacturers who are cooking their own soup.
Honarbacht: I agree with you. That’s why we invested a lot of energy in standardizing Zigbee 3.0, with the goal of tearing down the walls between these walled gardens. Ultimately, though, it’s not the technology that’s failed, it’s the business models. Restrictions are imposed on customers to protect products and costly developments. This is understandable from a manufacturer’s perspective, but a pain for end users. That’s why we have a different approach on our gateway. We don’t lock out third-party products, which is technically quite challenging. We try to support functionalities, regardless of the manufacturer.
And this will also apply to matter in the future?
Honarbacht: Exactly, a situation like today with some Zigbee bridges, which only pass selected lamps to Apple HomeKit, will not occur with us. What works on the Ubisys gateway will also appear in matter – as long as the respective product category exists in the matter standard.
“A situation like today with some Zigbee bridges, which only pass selected lamps to HomeKit, will not occur with us.”
Are you as actively involved in the development of matter as you are in Zigbee?
Honarbacht: We provide input, of course, but we are not a driving force in development. For various reasons. First, part of our business model is to license Zigbee software solutions to chip manufacturers. Matter doesn’t involve a similar business because the standard is open source. It also has a lot to do with Thread, a protocol that originally started as a Zigbee killer. It has potential; however, we feel Thread is not ready yet.
Where does the Thread protocol have some catching up to do?
Honarbacht: In energy efficiency, for example. Thread requires more energy for the exchange of the same user data than Zigbee, and with matter there is additional overhead. It is no longer enough to send a telegram; sessions have to be negotiated – encryption takes its toll. That’s why the energy consumption of matter products with thread is four to five times higher than with classic Zigbee. If we take the Green Power variant, which uses battery-less pushbuttons, for example, the difference becomes even greater. It remains to be seen whether such products can be realized using Thread.
However, Zigbee has had to contend with vulnerabilities in the past. Can the standard keep up with matter’s security mechanisms?
Honarbacht: Zigbee has a toolbox approach and has always been very scalable. You can implement it very simply, but also with a high standard of security, which corresponds to matter. In the UK, for example, meter readings are billed via Zigbee Smart Energy, encrypted from end to end. Zigbee Light Link is the opposite. A lamp can be reset virtually in passing by via the TouchLink function. The next Zigbee release, which is due at the end of 2022 or 2023, will eliminate this problem anyway – through new, modern encryption algorithms. Zigbee is far from beeing dead.
“Zigbee is far from beeing dead.”
Will existing Zigbee products be able to be updated to this next generation?
Honarbacht: That depends on the hardware. Older chips usually have a maximum of 512 kilobytes of program memory. Today’s developers have to juggle this to fit in additional code. Especially since part of the memory is needed to download the software update before installation. Many existing devices are not suitable for this. But in all our own models, including the first generation, we already built in data compression last year to ensure that they can receive full-fledged next-generation Zigbee firmware.
In addition, the whole thing is backward compatible. I can use the old products together with new ones. This is also a big advantage of a bridge solution: the possibility of gradual transition. Some companies have 800 Zigbee products in their range. Even if they wanted to, they can’t switch from one day to the next.
With enough memory, would it also be possible to switch to Thread later on?
Honarbacht: Yes, we are currently redesigning all products. The hardware can then do Zigbee or matter over Thread in the future. Or both together if necessary. Currently, I would say that Zigbee has proven itself very well in terms of the integrity of the network, even with very many devices from different manufacturers. So we see a long transition period. Some products in our range we have been maintaining for ten years. That’s not an unusually long period of time in building technology. And matter is a welcomed step, but not a panacea. The fact that there have been so many limitations in smart home systems up to now, that not every device worked everywhere, had nothing to do with the wireless standard, as I said. That was due to business decisions. And the business models don’t change automatically just because the matter standard exists.
Mr. Honarbacht, thank you very much for this interview.
Share this information: