Hacker News Reader

Comment by drmpeg | Hacker News Reader

This is a wacky idea, but the "modern" thing to do would be to repurpose seldom used 2-meter repeaters and make them 5G gNBs in the 3300 to 3450 MHz ham band (which overlaps NR band 78). Then you could directly use your cell phone (with a different SIM card) as your ham radio.

Network gaming servers to the gNBs and every kid in the US will want a ham license.

https://github.com/srsran/srsran_project

View on HN

I went down this rabbit hole for a while a few years ago. I think it's doable, but I don't have good answers to a few questions:

  1. How do you get cell phones to automatically identify in a way that satisfies the part 97 rules?
  2. How compatible are modern smartphones with the null encryption cypher?
  3. With fewer and fewer physical sim slots, is it possible for an amateur to produce valid eSIMs? (And is it possible without internet access, such that you could onboard new users during an internet outage?)

As other commenters mentioned, you need a core network too, which means you probably also need an 802.11-based backend network like HamWAN or AREDN plus some servers.

1.- You get a "nice PLMN" for the network and assign phone numbers (and manage key material) only after Ham Identification. DMR Style (at least here in Spain)

2.- Most of them will refuse to join any Private SA network that does not use SUCI and encryption (in my experience). If you have a "nice" PLMN, they will connect. What a "nice" PLMN is varies between modems, the testing PLMN is a safe bet :)

3.- Also gone down this rabbit hole. eSIMs must be signed by a GSMA authorized key, and they are picky. Osmocom people have relationship with someone that signs profiles for them (for pay). The crypto is quite straight forward to do offline, eSIM profile distribution not so much. Key propagation from generator to core network and final client (ue) would also be a challenge. Another option is using the testing certificate, it should remain active in most modems for certain PLMNs. I have yet to test this, but work gets in the way

There is a variety of open software and COTS hardware that could fully power this network. The client side is the hardest part, especially if you want a phone.

PC modems on a Linux machine are more manageable (But expensive, let's wait a few iterations on RedCap maybe). And SIMs with ADM keys can be purchased from different sources

I would love to set up a real managed digital network for hams, with stuff like roaming and the like.

But I'd be more interested in technologies that are one to many rather than the one to one calls on mobile phones. Open public channels keeps the ham community connected.

Something like DMR Tier 3 or Tetra would be great.

How do you "repurpose" a 2m repeater for that?

Swap out all the RF hardware. It's the high level site that matters.

You also need a 5G core to make that work. Also probably an IMS to actually provide some services on top of that. You also need to rip out all the encryption for the ham bands.

You could always somehow make the vibration motor a "movement feature", and call the phone a "remotely controlled vehicle"

;-)

Chatgpt- Yes, using the vibration motor to make small position changes on a remotely controlled vehicle would fall under the category of controlling the vehicle. If you encrypt the control signals that dictate when and how the vibration motor activates to achieve those position changes, it would be allowed under the FCC rules for encryption related to remote vehicle control.

Since these encrypted signals would only be used for the vehicle's movement or positioning, this approach aligns with the regulations permitting encryption for controlling remote vehicles. Just ensure that any non-control communications remain unencrypted to fully comply with ham radio rules.

I'm 99% confident that wouldn't fly and the definition of encryption is broad enough to cover even a "loophole" like this. I believe you'd have to demonstrate that you're actually "controlling" the phone rather than just performing communication and I'm not sure the communication with a vehicle actually is covered by the encryption cutout. My understanding is that it is so others cannot take control of the vehicle you're operating (which may leave you liable for whatever damage they cause with it, or at least makes that more ambiguous)

The real core is giant mess of servers but it's all software, no ternary vacuum tube device is involved. There are few OSS implementations and some even commercial ones that runs on single PC.

But why swap it out then? Just install the new hardware alongside.

just guessing but maybe to handle TDMA + QAM that ham gears don't support?

I think you misunderstand my point. My point was: If you have to replace everything why not leave the existing repeater up and add all the new hardware as well :) There is no need to take down the 2m unless the location is severely power or space constrained, which is not really the case for most of the high sites I've seen.

Replacing it only makes sense if you could reuse some hardware so you have to choose one or the other (or to invest in buying new). But in this case it will be 100% new so that discussion doesn't come up.

if it requires swapping out "all the RF hardware", why not keep the 2M repeater at this high level site and add your new repeater adjacent to it?

This is something I've thought about quite a few times, and I think it is a crazy idea but worth exploring. I'm way too naive here and have had almost no time to experiment due to other priorities.

But I think you could set up a global mesh network. I think what sucks is the encryption rules, but if there could be a big enough force that demonstrates it and its usage, those rules could get changed (there is already pressure to do so).

Now one thing though is there's some fuzziness about the definition of encryption with the FCC's definition[0]. Would this include routing protocols and/or compression? If they're open source protocols? If call sign is sent uncompressed and clear? This seems to be kosher considering other projects I've seen. So the messages might be able to be snooped on by anyone, but this would mostly be for fun and demonstration, like Ham is. But also, I think it would be pretty cool to have an open global communications platform. It would definitely also serve as a useful thing for natural disasters. (But we're not in the cold war anymore... we have encryption over the internet and that's even much more obscured. It would be nice to see encryption and a community driven global communication network)

The other thing I've been wondering is if anyone has tried satellite bouncing with the starlink sats? I'm sure they wouldn't be too happy about that, but hey, it is at least interesting to understand conceptually since you could really get a much larger view area and you don't need to travel all the way to the moon. But doing this you could potentially open up the network for coverage over oceans (at least more easily)

[0] https://www.ecfr.gov/current/title-47/chapter-I/subchapter-D...

Reply on news.ycombinator.com