DECT NR+: A technical dive into non-cellular 5G | Hacker News Reader

DECT NR+: A technical dive into non-cellular 5G

devzone.nordicsemi.com

86 pointsteleforce30 comments

From a few days ago, "What is DECT-2020 New Radio (NR), and how big a deal is it?":

* https://news.ycombinator.com/item?id=39849335

Same OP, same linked domain. Interesting. This blog post is significantly more thorough than the last one they posted.

Nordic Semi is a vendor in this space - many hardware vendors will make similar posts on different levels of technical depth in order to both drum up interest and build awareness of their products.

Note that the article has links to two modem chips that handle IIoT 5G radio links, including DECT NR+, that are made by Nordic Semi.

Content marketing at its best.

Smart move calling what is effectively an entirely new standard "DECT" to get free use of DECT's old frequency bands that are barely used by their original use for cordless phones anymore...

The bands themselves are unlicensed; anyone can use them. You don't have to be specifically called "DECT" to use them.

You do in some countries...

What countries require licensing for ISM bands?

DECT uses 1880–1900 MHz in Europe, other countries use similar frequencies, that's not an ISM band.

Fair enough, but that doesn't change the question significantly - what countries require licensing on DECT bands?

Most countries do require the equipent to be licesed, even the folks who operate and deploy then don't need to be.

And equipment for 1890MHz is not approved if it is not DECT.

DECT is still used a lot in factorys even today. It's way faster to type a short number, usually between 100 .. 999. The often used contacts you know anyway and you don't have to search in contacts. If the phone drops, nobody does care. DECT phones on work are great.

Ask HN: Why no mobile phone can also be used as a DECT phone? https://news.ycombinator.com/item?id=20909535

This can be a game-changer for certain low bandwidth unlicensed applications where the ongoing cost of cellular or satellite service makes the application economically infeasible. I could envision a whole layer of startup opportunities based on this technology from commercial applications like pets, construction, fleet management, security, and agriculture to a gazillion defense and intel applications. If I were younger, I would definitely dig into this further and compare it to LoRA and other existing radio technologies. Cheers!

This seems to be an order of magnitude better than LoRa (https://lora-alliance.org/ not https://arxiv.org/abs/2106.09685). LoRa doesn't have all the features this one does like OFDM, TDM, FDM, and HARQ. I didn't know there's spectrum dedicated for DECT use.

DECT-NR competition is 802.11ah and BLE Long Range. They all provide moderate bandwidth and moderate range, up to 1-2km. I bet they aren't low enough for battery-powered sensors, but probably good for battery-powered devices like Wifi or LTE.

My guess is that winner will be whoever get cheap devices out, and leverages their ecosystem.

Unless I am mistaken. Is that really the case? Both 802.11ah and BLE Long Range operate at lower frequency and also lower bandwidth, higher latency etc. DECT-NR could do everything the two could but not the other way around.

While 802.11ah does operate at sub-1GHz, BLE Long Range seems to be 2.4GHz, just with much more robust coding.

DECT-NR on 1.9GHz wil be lower in frequency than that and on less shared spectrum.

The comparison is not the same everywhere, as the European 868MHz bamd allows much less power (25mW vs. 1000mW) and much less bandwidth (5MHz vs. 25MHz) than USA 915MHz.

Similarly the USA 1.9GHz DECT band allows less power (100mW vs. 250mW) and less bandwidth (5channels vs. 10channels) than the European band.

So even before meshing DECT-NR has some major advantages over 802.11ah, in Europe.

LoRa works great because of the chirping it generates. Does DECT NR+ as well?

Why limit mesh topologies to cluster-trees? There are plenty of more efficient ways to route data...

I'd design it having message flooding for peer discovery (no hop limit, but a bandwidth limit - never use more than 0.1% of the total throughput for flooded messages).

Then, once connections are established, pick a few best routes, and send some proportion of data down each. Weight the paths via a cost function that takes into account load of each node, power use/availability of each node, impact of each flow on other flows (prefer getting nodes to transmit who cause least interference to other flows), etc.

Over time, adjust proportions of data down different routes to minimize the cost function.

This is what Wirepas is doing with their mesh network. They have a DECT version coming soon

Is this an unauthenticated mesh? Ie. might my neighbours device be forwarding data for my device and vice versa?

I want to see more unauthenticated mesh protocols.

By allowing 'unauthenticated' meshing, the total radio throughput in a typical urban environment is dramatically increased. By 2x or more often. Typical packets will take more hops at much reduced transmit powers each time.

The main reason not to do so is "what if my neighbour has crappy devices and black holes all my packets".

But your neighbour can already jam the whole spectrum and block all your packets. We design devices to meet specifications for a reason - and if the spec says "you must forward all packets according to this spec", and you mod your device to blackhole your neighbours packets, then the FCC will consider that jamming and treat it the same.

> FCC will consider that jamming and treat it the same

So... they'll do nothing?

yes.

But companies will be scared into only releasing devices that meet the spec if there is a decent risk that non-spec compliance leads to the whole company having their imports blocked.

Besides, forwarding packets for another wifi user will likely all be handled in the wifi chipset, so there will probably only be ~10 implementations by the ~10 companies worldwide who design wifi silicon. And if you're splashing out tens of millions of dollars on a wifi silicon design, you probably are going to make some effort to getting it sufficiently spec compliant to not be banned.

Not an unauthenticated mesh. This will be up user implementation iirc

Who administers the network? How is it monitored? This has the complexity of a small cellular network.

Does there exist DECT NR+ equipment yet? I imagine you can use the same RANs for this as "5G", but those are ridiculously expensive. What about mobile devices?

TFA mentions this[1], but it's not IoT cheap.

1: https://www.mouser.com/ProductDetail/Nordic-Semiconductor/NR...

Not that expensive for the space it's used in.

I meant, more integrated boards.

Great! Another thing that can be bashed into the term "5G" for marketing unrelated technologies.

Reply on news.ycombinator.com