By John WB2OSZ —

Build a High-Speed Data Exchange HamNet Based on APRS-IS

by fangzhou.xu BG5ESN

We all know that although APRS provides a global short‑message network, its rate limitations and broadcast constraints prevent it from becoming a true innovation platform for worldwide amateur radio digital infrastructure.

For this reason, I propose a brand‑new high‑speed global data exchange network for amateur radio operators, built on APRS‑IS. The idea is remarkably simple: use APRS‑IS as a discovery directory, and use additional data servers to carry the actual data.

APRS‑IS already offers a presence broadcast for amateurs around the world. We can leverage this broadcast to build a presence‑based server directory. If we can ensure that these servers are used only by licensed amateur radio operators, then we have created a high‑speed, integrated data network serving the amateur community.

This design uses only the basic capabilities of APRS‑IS. The high‑speed data part is fully handled by any server discovered through that directory. This layered approach not only reduces traffic on APRS‑IS, but also allows amateurs to exchange data at much higher speeds.

The engineering implementation is very straightforward:

  1. Use digitally signed APRS packets so that any data server provided by anyone can be discovered by other amateurs.
  2. Amateurs verify the originator of the packet and add that server to their own directory.
  3. Amateurs then exchange data at high speed through that server.
  4. When the server operator stops broadcasting, or when other amateurs no longer receive the broadcast for an extended period, the server entry naturally disappears from the directory. If the operator also shuts down the service, one round of high‑speed data exchange is completed.

Although the engineering description is simple, there is a socio‑engineering challenge: how do we get everyone to openly trust a data server provided by any user? We would like to believe that the world is good, but in reality an open server is hard to guarantee for HAM‑only use.

My solution to this is: the developer issues a hardware‑based identity credential. This hardware is bound to a real amateur radio operator’s identity, so that the user’s callsign is verified at the edge. This avoids most problems. All amateurs can then conduct trusted data exchanges under the same authenticated and trustable conditions.

Each hardware node can autonomously filter trusted APRS packets and build its own server list. Signature verification is built into every standalone device.

A special note on the dual role of the hardware node:

  • Role 1 (Client / User): It acts as an end‑user device that verifies the identity of other servers and uses them for high‑speed data exchange.
  • Role 2 (Server Announcer): When a data server does not have APRS transmission capability itself (e.g., running on a cloud VM or home broadband), a nearby hardware node can broadcast that server’s basic information onto the APRS network on its behalf. In other words, the hardware node is not only a trust anchor but also a “loudspeaker” that helps third‑party servers appear in the directory.

These two roles can exist independently, or the same hardware node can perform both.

There is also the question of who provides the data servers. My solution directly leverages the second role described above. Any hardware node can broadcast on behalf of a data exchange server, allowing that server to be discovered by amateurs worldwide without needing to connect to APRS itself.

This means that the entire data link can be established using only a set of trusted hardware. When an amateur announces a data server, other amateurs add it to their directories and use it to exchange data with one another.

Thus, any amateur with server resources can temporarily set up a server, and any user can leverage their trusted hardware along with these server resources for high‑speed communication. This breaks away from the old logic of “build a network first, then exchange data” and replaces it with “dynamically create a data exchange server only when data actually needs to be exchanged.”

This is the fundamental idea behind my design of FMO (NFM Over Internet). FMO is an amateur radio data and voice exchange network built on this architecture, following the same design philosophy.
If you are interested in this project, please visit my website:

https://bg5esn.com/docs/fmo-product-guide/

73

fangzhou.xu BG5ESN