Jim Lux said: > What I'd like to do is have a way to transmit an "in-band" signal that > can be used in post processing to precisely synchronize the data streams. > Ideally to, say, 1 ns. > Conceptually, this could be something like taking a 1pps from a GPS > receiver (with whatever uncertainty it has), or a GPSDO,� and generating > a synchronized tone burst that's in band. So the question is, what off > the shelf box, or combination of inexpensive boxes, could generate such a > signal. I asume you want to generate a signal that it something like N cycles of a reference clock. Use your reference clock to drive a PIC. Feed the PPS in on a GPIO. Wait for PPS, turn on an enable signal, count to N, turn it off. If your reference signal is too fast, you can build that recipe in chips or an FPGA. Don't forget the synchronizer on the PPS. Or divide by 2 to run the PIC at half speed and count to N/2. That gets you a signal you can find, but you don't know the delay from the PPS to that signal. So run another SDR setup located where you are making the signal and feed the PPS directly into that setup. That will give you the offset from the PPS to your signal, and use that offset to add another layer to your clock fixup software. -- These are my opinions. I hate spam.