[time-nuts] time-nuts Digest, Vol 168, Issue 21 SDRs and precision time
jimlux at earthlink.net
Tue Jul 24 14:07:16 EDT 2018
On 7/23/18 5:58 PM, Graham / KE9H wrote:
> It would be a lot easier if you used a GPSDO to synchronize your SDRs,
> since both the time and frequency derived are several orders of magnitude
> better than what you can get out of an LF receiver. Any systems like this,
> that I am aware of, use GPS, not LF.
This is essentially what we are doing for SunRISE - we're going to be
flying 6 small satellites in a cluster about 10-15km across to image the
sun from 0.2 to 22 MHz (or thereabouts).
We have a SDR which digitizes the RF, and a GPS receiver, and we log
both the GPS observables and the filtered/downconverted RF. Then, on
the ground, we reconstruct the time and position when the measurements
were made to do interferometry.
On Earth surface, this would be somewhat easier - I've been fooling with
using 4 RTL-SDR receivers, 4 beaglebone green wireless, and 4 GPS
receivers to do something similar.
As it happens, the challenge is in the non-determinate nature of the
data path from RTL-SDR digitizer via USB (on the space version, we can
arrange to use the same OCXO to drive the GPS sampler and the HF
sampler, etc.), so I'm working it by generating a pilot tone that's
Some years ago, we build a breadboard to demonstrate that this kind of
thing is possible for a large array at L-band.
More information about the time-nuts