NPR-70 IPv4 Radio Modem | New Packet Radio Over 70cm Band 

Do a search online for RSGB VHF Manager report (latest 2021?) and read what they say about NPR. I'm not a fan of the RSGB, I'm sure the people in general are fine but I feel my licence was issued by OFCOM (as it is now) and we need cooperation not managing. I got a msg to say (amongst other comments) that they would not be issuing NOV's for this project. Surprised me, as I though the NOV's were issued by OFCOM and not the RSGB? Perhaps it's just me though ;)

dont know your location but in the US encryption via your ham license is illegal

@@christopherbrown3198 It's not encrypted as so many have replied to me. The problem is the speed and bandwidth. Not part 97 compliant.

In the UK, Raynet are allowed to use encryption at the request of the User Services which this is. However, I'd probably have to get them to specifically ask before using it ...

@@christopherbrown3198its always puzzled me that even though those days you can pretty much encrypt whatever u like with your cellphone, however all the rules for ham haven't changed since the days that encryption was considered a top secret military thing 🤷‍♂

That job is probably better done with LoRa transceivers, which are inexpensive and pretty robust over longish distance, though with pretty slow data rates. I and another ham in my town have experimented with Meshtastic, a mesh protocol designed around LoRa modules operating in the ISM bands. We managed to pass messages across about 6 miles of mesh nodes with the longest single hop being a little over 2.5 miles in a mesh of 5 nodes total. Meshtastic modules can be had from $20 to $35 each.

@@Randrew What kind of bandwidth do you get over LoRa? I just started playing around with it. I think both of these have their strong points. This modem system being IPv4 compatible is a BIG plus for me. You can do so much more than message and share your GPS location.

@@gingerman5123 LoRa bandwidth can vary wildly from a couple hundred bytes per second to maybe 50 k bytes per second. Depends on the distance and modulation method(s) chosen. LoRa is good for sensor networks and low rate messaging, but not great for general Internet.

Good question. What say Tech Minds?

When you're comparing this to older, existing amateur data modes, those ping times do seem out of line. Consider old-school packet radio twaddling along at 2400 baud... an IP ping packet of 84 bytes or 672 bits would take 280 milliseconds just to be sent. Double that for a reply. But this box is supposed to be jogging at ~65 kbps so transmit time should be negligible in comparison. I too would be interested in ping comparisons between modes 20 and 24 because if there's no significant difference, then the overhead probably lives only in the software. I did note that the design is based primarily on 4 off-the-shelf modules: a DC-DC converter, W5500 Ethernet interface, a STM32 Nucleo CPU stick and an RF4463F30 transceiver. Quick research of the W5500 network module indicates it should NOT be the cause of high ping times here, so for now I'm going to blame the code in the Nucleo board. Or... maybe the power level TM set is just too low or there was some other unexpected signal problem.

@@Randrew I bet you're right about the software, or the signal even. If there wasn't enough loss between the two nodes this could also ruin performance.

@@Randrew I am the designer of NPR-70. The high ping is due to the protocol. It varies depending on the modulation parameter. Modulation 24, ping is arround 80ms. Modulation 20, ping is arround 500ms, in "high speed mode". Tech-Minds was in slow mode.

@@F4HDK Thank you for the reply! When this video first came out I did some cursory research, but now I've lost where I found protocol information - which I didn't study in detail. I admit my interest waned some when I found it wasn't fully legal in the US ;) My tentative technical take-away is maybe IP encapsulation on modern low-power transceiver designs is not going to be efficient, speed-wise. BTW, I do really appreciate your modules design approach. It takes advantage of proven commercial designs AND their economy of manufacturing scale.

technically yes

lol, give me two of the radio and ill do it. hahahhaa

yes, and just read manuel, you will got answer

any 70cm band Yagi or something. its a must to get some range out of this.

It is entirely possible but check part 90 regs. They have allowed so much more for GMRS than when I was active with it. I still have a callsign but I don't see the point where I live now.

@@W4TRI I'm not sure the device can be tuned that easily. My understanding of the TRX module they use is you need to order it for one of 5 frequencies: 315, 433, 470, 868 and 915 MHz. I guess these all correspond to ISM bands available in different regions of the world. The Si4463 chip itself can be programmed to operate from 140 to 1050 MHz, but the board design and component selection seems to be limited to those frequencies I listed. And this black box has the 433 MHz version. The 470 MHz version *might* could be bent down ~10 MHz to FRS/GMRS by changing the software config... but that's not the version in use here. On a slight tangent... all those 433 MHz ISM devices I see chirping all the time don't even belong in ITU Region 2 (North America). Everyone here with an Accurite wireless temperature sensor should have a amateur radio license... but then they'd be transmitting out of compliance anyway :|

Hello. You cannot use this NPR-70 with 433MHz ISM rules. The power is way too high. ISM is restricted to 10mW, at output of antenna, with such a large bandwidth. You would get a range of ~100 meters with this low power.

Something like that would be better suited for the 915mhz ISM band as it's much more lenient with power and bandwidth

​@@LunaWunalooking at the datasheet for the rf module used RF4463F30 looks like there is a 915mhz version available, which should be pin compatible as well. since afaik the project is open source, should be a simple task.

This device is both the radio and the modem together, no other radios needed or accommodated. And regular walkie talkies don't pass enough bandwidth for even the slowest modulation scheme of this device - 100 kHz.

yeah basically as far as i can tell, the issue with walkies talkies is the fm de-emphasis filter on the receiver side is too narrow for wideband digital, so only 1.2kbps or so.. you can try skuep/AIOC with wb2osz/direwolf for some slow digital comms with your walkie talkies, otherwise just use them for what they meant to, and this modem for the digital data comms.

They are legal, And it Seems that you can use this in the EU on the 433.42 Mhz ISM band on 1MHz bandwidth without issue. So no license required in its Stock Configuration.

@@henrikkristiansen3869 the 500mw isn't, iirc depending on the country you have to run it at 10mw

Could .....yes, but a No-Go for ham. Amateur radios very open and any encryption is strictly prohibited.

@@tellnets Yep and these are totally illegal for US hams anyway. Closed network that can't be monitored/decrypted.

@@W4TRI These are legal for US Hams, Just no encryption

@@christopherbrown3198 No they are not. No way to decipher. And this is a closed loop system.

@@W4TRI This is just another digital protocol. Look into HSMM. As long as we comply with Part 97 rule, and the purpose of the encoding is not to obscure the meaning of the message, we are fine.

Yeap. Layer 2 bridge would have been nice, looks like some proxy ARP bodgery to get network working.

​@@steve87uk I think a pure L2 bridge may be too simple to work well. Modern networks are a bit too chatty to "just work" with half duplex and (in today's measurements) low speeds. I do think it needs to be smarter to avoid roundtrips and random broadcasts and probes. So I think some layering violation may be necessary. But I think this could have been done with IPv6.

IP stacks of modern OSes require high speeds, low packet loss and low bit error rate. TCP of modern linux or Windows host will quickly choke up when latency is high or the packet loss is over 14%. QUIC is even more resource hungry, compared to TCP. So, in case of ham band modems, I would prefer to have an AX.25 implementation, maybe with features like Hybrid ARQ.

Literally only a very small number of the world's ISP's support ipv6 fully. Almost all will support IPv4 for at least 20 years either directly or using a feature in routers to allocate a virtual IPv6 address to legacy devices on your home network. Even when legacy devices need to work on IPv6 that doesn't support them, you can use a dedicated router to add this support, even if this means using 2 routers in your home network (something you should do with many IOT devices anyway for full security). There are too many legacy devices around the world that do not understand IPv6 for a 100% IPv6 network to exist for decades yet.

@@dennis8196 you missed my point about the *future*. And it's not a very small number of the world's ISPs. Google has, what 25% of all traffic IPv6. And it's not like you can use NPN for internet anyway, since who would use internet without encryption? The world will also continue using fax machines for many years (my bank requested it months ago), but I would not invent a "new" framework that would include fax machines.

it is not ionizing. so, no.

@@RoelandJansen I assumed he must be posting his love for his favorite death-metal band.

Lack of knowledge is cancer and makes Q-anon and other retards thrive. You will not get cancer from radio waves unless there are insane power levels directly to your body. The size of radio waves compared to the electrons in our body protects us. That is why you can get skin cancer from regular sunlight, but not radio cancer. I have climbed towers with live 250.000W pulse effect turned on. Still no negative effects from that.

go visit chernobil 😉