The 3:17 AM Signal That Wasn't There Yesterday — Or On Any Frequency Chart

03:17:00 — Six nights in a row now.

I run a pretty comprehensive monitoring setup. Software-defined radio, spectrum analyzer, multiple antennas optimized for different bands. I log everything automatically — have been for years. The system captures signal strength, frequency, modulation type, duration. It's not fancy by professional standards, but it's thorough.

Six nights ago, I was reviewing the overnight logs and noticed something I'd missed in real-time. A carrier wave at 164.375 MHz. Strong signal — minus 42 dBm at my receiver. Duration: exactly 47 seconds. Start time: 03:17:00.

That frequency shouldn't be in use out here. It's in a band allocated to federal operations, but the nearest federal facility is 90 miles away. Signal strength suggests local origin — maybe 5-10 miles maximum.

I started watching for it.

Night two: 03:17:00. Same frequency. Same duration. Same signal strength within 2 dB.

Night three: I'm at the equipment when it happens. The spectrum analyzer shows the spike. My SDR software displays the waterfall. But here's what bothers me — when I try to capture the actual signal for analysis, the recording shows nothing but noise floor. Equipment records the *presence* of the signal. But not the signal itself.

I've checked my recording chain three times. The hardware's fine. The software's fine. I can record everything else that comes through — shortwave broadcasts, weather satellites, amateur radio operators, aircraft transponders. Just not this one signal.

Last night I drove toward it.

I've got a directional antenna setup I can run mobile. Takes about 20 minutes to get a rough bearing on a signal if it holds steady. At 03:15 I was parked at the crossroads near Old Miller Farm with the antenna mounted on the roof rack. Signal came through at 03:17:00 exactly. Bearing: roughly northeast. That puts the origin somewhere in the national forest land — the same area where The Dead Zone's been migrating.

I started driving that direction. Signal stayed constant for the usual 47 seconds, then dropped. By the time I got two miles up the forestry road, there was nothing but crickets and the usual overnight chatter on the ham bands.

The mundane explanation is right there, and I know it.

Federal agencies conduct encrypted communications tests. They use automated transmitters on precise schedules. The fact that my recording software can't capture it? Probably something about the modulation type — spread spectrum, frequency hopping, something my civilian gear isn't equipped to decode. It shows up as a carrier wave but the actual content is beyond my equipment's capability.

The location in federal forest land? That's exactly where you'd expect test equipment. The precise timing? Automated systems. The 47-second duration? Programmed parameter. The numbers don't lie, but they don't always tell the whole truth either.

But here's what doesn't fit cleanly:

This signal wasn't in my logs a week ago. I keep seven years of records. I've got automated alerts for new signals in this band. Nothing. Then suddenly, six nights ago, it appears. Same time. Same frequency. Same duration.

And when I physically moved toward it, it stopped exactly on schedule — not early, not when I got close. Just... 47 seconds and done, same as always.

I've been tracking this for 7 years now. I know the difference between wanting to see something and actually seeing something. I know how easy it is to pattern-match coincidence into conspiracy. But I also know my equipment, and I know this area. And something started broadcasting at 03:17 AM last Tuesday that wasn't broadcasting the week before.

Tonight I'm setting up a second monitoring station at The Ridge. If this is local, I should be able to triangulate a more precise location. If it's a test, it'll probably run for a standard duration — two weeks is common for this kind of thing.

Another entry for the log.

What I'm asking:

Anyone else monitoring VHF federal bands? Anything unusual in the 164-165 MHz range at odd hours? And here's the question that's keeping me up: why would an automated test system start exactly at 03:17 AM? That's not a standard interval. Not midnight. Not on the hour. Not even a clean 15-minute mark.

17 again. Always 17.

Stay vigilant.