Remote station design: control paths, latency, security and the rules (IC-7610/RS-BA1, SCU-LAN10, Flex, RigPi, Kiwi/WebSDR)
- Replies 2
- Views 43
- Created
- Last Reply
A better way to browse. Learn more.
A full-screen app on your home screen with push notifications, badges and more.
Remote operating has two parts that get mixed up: the radio's control path and the station's safety and legal wrapper. Most threads jump straight to "which box should I buy" and skip the second part. This is a framework to argue about, so please poke holes in it.
The first question is whether the radio itself speaks network. The Icom IC-7610 has an Ethernet port and works with Icom's RS-BA1 server/client software, so the radio does CAT, audio and (via the software) keying over the LAN. Open-source clients such as wfview target the same Icom network protocol. Yaesu's SCU-LAN10 is a separate network unit that bridges supported radios to a LAN, so check Yaesu's compatibility list for your exact model and firmware before assuming it works. FlexRadio is different because the radio is the SDR. SmartSDR clients talk to it directly, and SmartLink handles access over the internet, so there is no separate audio bridge. If the radio has no network path at all, the usual answer is a small always-on computer at the station running Hamlib/rigctld, a sound interface and a remote desktop or audio-streaming layer. A RigPi-style Raspberry Pi build fits that role: the Pi sits at the radio and you connect to the Pi, not to the radio.
Where the software runs matters a lot for digital modes. For FT8, FT4 and similar, run WSJT-X at the station and remote-desktop into it, rather than streaming audio to a client at home. Those modes depend on accurate computer time and on clean audio levels. Both are easier to keep right locally, and network jitter then only affects your screen, not your decodes.
Latency also differs by mode. For SSB, round-trip delay builds up from the audio codec and buffers, the internet path and any jitter buffer. Once the total gets long enough to disturb turn-taking, it feels like a satellite link. CW is harder, because streaming key-down/key-up over a variable-delay link can mangle the timing. It is generally better to send text or keyer commands to a keyer at the station, or to use the radio's built-in keyer, so the timing is generated locally. Adding VPN and remote-desktop layers on top of audio streaming adds delay at each step, so measure with your actual stack instead of trusting a spec sheet.
On security, I'd treat port-forwarding a radio server straight to the internet as the thing to avoid. Radio network protocols were not designed as hardened internet services. A VPN, such as WireGuard or OpenVPN, or a managed overlay network, lets you reach the LAN without publishing radio ports. Beyond that, keep the radio and Pi on their own VLAN or at least their own subnet, turn off UPnP on the router, use unique credentials, and keep router and OS firmware current. If a vendor's remote feature needs inbound ports, read exactly which ports it needs and what authentication protects them.
Fail-safe behavior is the part people forget. What happens if the link dies while you are transmitting? Use the radio's time-out timer, and add a watchdog or relay that drops PTT when a heartbeat disappears. A remotely switched power strip lets you recover a hung PC. Also think about status feedback for antenna switch position, amplifier state and SWR, since you cannot walk over and look.
On the rules, remote control is generally permitted, but the details depend on your country and licence class, and I'm not going to cite a rule as universal. In the US, remotely controlled stations fall under the Part 97 station-control provisions (control point, control operator, protection against unauthorized control, limits on transmitting if the control link fails). Read the text yourself, starting with 97.109 and 97.213. Operating from or through a station in another country raises separate licensing questions. Contests and awards also set their own remote-operation rules, for example where the station and operator may be located, so check the sponsor's current rules before relying on a remote QSO.
Receive-only networks are useful as an extra tool. WebSDR and KiwiSDR sites let you check how your signal sounds at a distant location or confirm band conditions before you fire up the remote rig. They are not a real-time monitor of your own audio, because the stream delay is large. For services like RemoteHams.com, which offer shared or hosted remote access, read the current terms on licensing responsibility and who counts as the control operator. I'm not going to assume how any particular service handles that.
What does your control chain look like, and which piece has been the weakest link: audio quality, CW timing, network reliability or the security setup?
Link to comment
https://www.hamradiobase.com/forums/topic/10867-remote-station-design-control-paths-latency-security-and-the-rules-ic-7610rs-ba1-scu-lan10-flex-rigpi-kiwiwebsdr/Share on other sites