There are a few options if you want to network computers on amateur radio. There are WiFi hacks of sort, and of course there’s always packet radio. New Packet Radio, a project from [f4hdk] that’s now on hackaday.io, is unlike anything we’ve seen before. It’s a modem that’s ready to go, uses standard 433 ISM band chips, should only cost $80 to build, and it supports bidirectional IP traffic.
The introductory documentation for this project (PDF) lays out the use case, protocol, and hardware for NPR. It’s based on chips designed for the 433MHz ISM band, specifically the SI4463 ISM band radio from Silicon Labs. Off the shelf amplifiers are used, and the rest of the modem consists of an Mbed Nucleo and a Wiznet W5500 Ethernet module. There is one single modem type for masters and clients. The network is designed so that a master serves as a bridge between Hamnet, a high-speed mesh network that can connect to the wider Internet. This master connects to up to seven clients simultaneously. Alternatively, there is a point-to-point configuration that allows two clients to connect to each other at about 200 kbps.
Being a 434 MHz device, this just isn’t going to fly in the US, but the relevant chip will work with the 915 MHz ISM band. This is a great solution to IP over radio, and like a number of popular amateur radio projects, it started with the hardware hackers first.
I actually found it rather impressive in the project page, seeing someone revamp this rather obscure protocol of packet radio.
Thanks for sharing.
Hello,
this is not a respin of old packet radio.
This is a completely new protocol that just transfer IP frames.
There is no HDLC, no UI frames, no G4RUH scrambling, nothing like that.
Just a si4463 radio, a new TDMA frame format, and IP.
Would it be easy to make an “FCC-friendly” version of this that offered modulation options with 56k symbol rate and <100 kHz bandwidth? I think 20 kHz deviation for GFSK, 10 kHz deviation for 2-GFSK and 5 kHz deviation for 4-GFSK ? If I understand, with 4-GFSK the raw data rate would be 214 kbps even though the symbol rate is 56 ksps.
434MHz is in the US amateur band I think. In any case, this chip can tune a quite wide UHF range so you can certainly select a frequency within the band.
I’m F4HDK, the designer of this “New Packet Radio” modem.
Of course, you can select the frequency. Currently within 430 – 440MHz range.
It’s not an ISM solution, it’s really an amateur-radio solution, which uses ISM chips.
And I don’t know why Brian mentioned a datarate of 200kbps… You can achieve 500kbps (net, effective datarate) at max.
Last, The Hamnet network, here in Europe, is not designed to be connected to Internet.
500kbps is plenty of speed. Your project in a mesh would be a nice test for the matrix protocol. Not for people to chat but for resilient arbitrary data (environmental sensors, etc) transmissions.
https://matrix.org/blog/2019/03/12/breaking-the-100bps-barrier-with-matrix-meshsim-coap-proxy/
DK1DB
The whole idea of using ham radio as a means of ip connectivity is awesome to me. I always thought with wifi being such a small area of service in most cases they could take a page or two from the ham operations category. I guess it would require substantial change to hardware and equipment to make long range wireless ip work for the masses. 200kbps isn’t that bad though.
My ham radio club uses 5 GHz amateur radio links between its repeaters. Pretty cool stuff, though we’re still working out the kinks.
As for the legality of this in the US: American hams have legal access all the way from 420 MHz to 450 MHz, unless they are above “Line A” near the Canadian border, in which case the allowed 70 cm band is from 430 MHz to 450 MHz instead. There would be no problem transmitting on 433 MHz. Of course, the maker of this project did say that it is frequency agile, so it’s a non-issue.
The issue in the US is that FCC rules for the 70 cm band limit data rates. Symbol rated is limited to 56 kilobaud and for “unspecified” modes, total bandwidth must not exceed 100 kHz. Really puts a damper on things. The rules do permit transmission of analog “fast scan” television in bandwidths up to 6 MHz on 70 cm. But not digital data. Here in the Northern Virginia area, the band is covered with government stations using 5 Mhz wide digital modes (that hams are not allowed to use).
Really a question for F4HDK –
If I wanted to define another modulation scheme it looks like I would modify the ‘radio_config’ file
// USER DEFINED PARAMETERS
// Define your own parameters here
// INPUT DATA
(/*
// Crys_freq(Hz): 30000000 Crys_tol(ppm): 20 IF_mode: 2 High_perf_Ch_Fil: 1 OSRtune: 0 Ch_Fil_Bw_AFC: 0 ANT_DIV: 0 PM_pattern: 0
// MOD_type: 5 Rsymb(sps): 500000 FdevHz): 41666 RXBW(Hz): 150000
etc
and use the radio_config_generation.exe program to make an SI_4463_config_xx.h file to use with the program
yes?
Ken
Nicely done project and really great data rates. There seem to be lots of trade-offs available on band, radio choice, bandwidth and power. This is a great demo for telemetry and remote control of ethernet devices. It caused me to also look at some of the longer range/much lower data rate variants like Winlink/ARDOP and even the SCS/Pactor modems. This is a great implementation where you take advantage of off the shelf radio chips without getting into the DSP physical link level. Keep up the hacking! Thanks for publishing – John/AG5PL
why not include normal linux? i need untp , email etc. how using without linux? i must power for second device?
why?
well, todays “ham” is very different from my era having grown up and worked operationally in all coast radio stations in NZ, overseas and on various ships – from 1958 onwards…this stuff is of little interest but, hope everyone else gets something from it…cheers
Brian
This device is a bridge, like a wifi acces point. If you want linux, you do it on the computer that is connected to this modem using an ethernet cable.
including linux on THIS device makes no sense, it would be completely overkill and power hungry. Use the raspberry pi for something else :)
Look s interesting, I may give it a go, How do you suggest going about getting the PC ?
Do you mean “getting the PCB”?
It depends where do you live.
For Europe, I will sell kits.
For somewhere else, you can send the “gerber” files to a PCB manufacturer, directly on their website. For example JLCPCB. It’s cheap : ~20$ for 10 PCB, plus shipping.
Could this be used with an ibiquity type system to have nested digital channels in a single analog signal?
What is the latency of this system? Thinking if this could be “abused” as a DYI control link for a long range FPV drone…
after skimming cursively through the documentation, i think the latency might be pretty high for this use case. you might get it down a few notches by changing some firmware stuff….
I like the idea, but the regulations in some countries might not allow…
You are right, but it does not matter. We can still do experimentations in places where we are allowed.
And this again proves my previous comments on similar articles. Silly and archaic regulations, sometimes rigorously enforced by retired hams (who apparently have FCC enforcement as their hobby for retirement) are huge barriers to using amateur frequencies, often rendering them borderline useless.
Notice that another ham mentions using 5ghz (most of which is an *ISM* frequency!) as their repeater backhaul?
And even if you can manage to get ARRL and its army of bored retirees out of the way, and get the FCC to remove the really silly and old archaic stuff like the 56k limit on data (but bandwidth wasting 6mhz wide streams of TV data is just fine…wth?) Your still left with the no encryption problem. Without some movement on the encryption issue most data streams become useless anyway because so few internet services function with unsecured data streams. How long before the focus on security renders modern browsers usless on a service like hamnet (non ttps browsing is already being depreciated in browsers).
Projects like this wither and die the second the founder moves on to something else (see FaradayRF) because there is essentially zero commercial gain in pursing it. Why work on a amateur radio project when an ISM project has a much larger user base and it has potential commercial potential (even if its only from the R&D)? Amateur Radio contributed a lot back many many decades ago, but that was way before the internet and the restrictions placed on amateur radio didn’t limit its crossover utility into commercial areas like they do now. And even more importantly, ISM didn’t exist in the form it does now.
Even as a general class ham I would oppose the addition of any new frequency being added to the amateur service and instead suggest it be turned into a unlicensed band. The frequency would be far more useful there then it would rotting on the vine in the amateur service (especially anything above 500mhz since 99% of amateurs basically ignore everything above 70cm).
@banp968
I agree with you, the regulations of apateur-radio are often outdated.
Here in France, we don’t have modulation limits anymore.
But IARU recommendations are also very very outdated.
Transfering this project to a pure ISM project would be useless, in my opinion. I will not be able to make links at several dozens of kilometers and 500kbps with the restrictions of ISM.
With amateur radio, you can use gain-antennas, and interesting radio power (20W here), without restriction.