Hacker News Reader

Comment by ricksunny | Hacker News Reader

Wondering if this will spur innovators' handoff-based mesh networks (slow, low-bandwidth but very, very democratic).

When whitespace in television bands went unlicensed i don't know how much of that we saw: https://www.fcc.gov/general/white-space

I feel like the barrier may be whether dedicated hardware is required or not. In such a large band, 6 GHz, I would expect a lot of generalized (i.e. non-dedicated) platform hardware to be developed & offered allowing software-focused innovators to offer into the long tail of applications, including mesh network(s).

View on HN

Everyone I've talked to from the LORA crowd and even some (LORA) alliance guys tells me lo-energy meshes are hard to get right. Am I missing something?

The new DECT NR+, the first non-cellular 5G standard supports mesh configuration but fixed not mobile network:

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

https://devzone.nordicsemi.com/nordic/nordic-blog/b/blog/pos...

I wonder if ESP-NOW[0] would be useful for this. I've been toying with building some mesh-based lighting controllers for synchronizing lights. Fortunately it can be entirely static (number of nodes) which makes it an easier problem than dynamic meshes.

0: https://www.espressif.com/en/solutions/low-power-solutions/e...

Correct. Source: I do low-energy mesh networks. There are loads of tradeoffs, and a lot of it just has to do with the physical medium.

Sure, you could go with an off-the-shelf solution, but those aren't always a good fit for whatever product you're developing. If you can squeeze your bandwidth requirements to an absolute minimum, say, tens of kilobits, you can accomplish absolutely nutty stuff with a cheap LoRa radio and a well-designed antenna. Sure, it's tens of kilobits, but it's tens of kilobits sent and received over a mile between devices with batteries the size of your pinky.

Tens of kb worked charm for me in the 90s. For iOt is more than enough. I’m more concerned whether the mesh works when devices move around yet still keep within vicinity of each other.

You nailed it. The overwhelming majority of research around mesh networks assumes that nodes are stationary, and connection lifetimes are long-lived and consistent (if not perfectly reliable). My job is extra cool because none of those assumptions apply.

This is fascinating! How can I learn more about this tech if I wanted to try an build something myself?

Mesh networks are a neat idea, but the reality of is often disappointing e.g. given compatible radios all the IoT devices in a room could form a mesh network of equal peers, but if a few lightbulbs use their position (mains powered, good location) to form a hierarchical network it will actually work (without wasting battery power and air time).

This feels very pie in the sky and dreamlike. Of course some corp will figure out something to do within this space and make closed products that push out free and open uses. At least that's my pessimistic view opposed to your optimism.

If you pop open a spectrum analyzer in most places, you'll see a ghost town in the Cathedral and a hopping lively Bazaar in 2.4/5GHz. On net, it is good that more resources are going to the Bazaar.

That's a Bazaar I would love to window-shop in - and heck maybe purchase a couple of ornate unique carpets while I'm at it =)

6ghz is pretty fragile -- i can't imagine one product "pushing out" another when a wall will block the signal. Just don't have the closed products in your home, and the open ones should work just fine...

Reply on news.ycombinator.com