Radar Horizon / Track Data Provenance (1992)

I passed one of the hardest exams of my career as an Ensign.

Four hours. Oral. I had to trace electrons at different frequencies — which antenna, which patch panel, which console in CIC. Air tracker. Surface tracker. DRT. Sonar. EW. Navigation. Fire control.

I passed. And because of that exam, I knew exactly what I could trust on every console in that Combat Information Center.

What I couldn’t control was what happened after my data left the ship.

It was 1992. First deployment. Transiting the Pacific, the South China Sea, the Indian Ocean. We had a C2 system called JMCIS that auto-transmitted our Link-11 tracks up the chain — all the way to CINCPACFLT — over the OTCIXS network.

I was comfortable with that setup as long as the track was inside my radar horizon. Inside that boundary, I could validate with 100% certainty that what we were reporting was accurate.

But there was no auto-stop when a track passed beyond my SA.

Those tracks kept transmitting. Floating at whatever course and speed my system had last tagged them with. Long after I had no idea where they actually were.

Somewhere up the chain, a flag officer was looking at a tactical picture that included data I could no longer vouch for. Data that looked authoritative. Data that was stale the moment it crossed my radar horizon.

I was 23 years old and I already understood the problem that still hasn’t been fully solved 33 years later.

You might think AIS changed this. In some ways it did — maritime tracks are now continuous and global rather than radar-dependent. But the validation problem didn’t go away. It just changed shape. Now the question isn’t whether the track went stale past your radar horizon. It’s whether the track was real to begin with. Spoofed positions. Manipulated identity signals. Vessels broadcasting one location while operating in another. NATO’s ongoing work on maritime track validation exists precisely because an authoritative-looking track and a trustworthy track are still not the same thing.

The system still doesn’t know what it doesn’t know. And the people at the top of the chain still have no way to tell the difference between what’s validated and what isn’t.

That’s not a technology problem. It’s an information architecture problem — and it starts with a question that no centralized system can answer on its own: who at the sensor said ‘go,’ and when did they stop being able to vouch for it?

It’s the same problem I’ve spent the last 15 years trying to fix at the Joint level. And it looks different in every domain — maritime, air, land — but the wound underneath is identical every time.