4 ms·

With synchronized receivers you could do some rudimentary direction finding. Note that synchronizing SDRs is much more achievable if they're physically nearby (

by mNovak 3y ago

With synchronized receivers you could do some rudimentary direction finding. Note that synchronizing SDRs is much more achievable if they're physically nearby (e.g. can run a cable between them for a common clock) vs if they're physically distant observers (can't exactly use GPS time for synchronization if you're measuring GPS interference)

toomuchtodo 3y ago
Very helpful (as well as sibling comment by pierat). Thank you both. I will need to do some research with regards to pairing SDRs with local disciplined clock that can tolerate temporary loss of remote (stratum-0,1) time precision. https://en.wikipedia.org/wiki/GPS_disciplined_oscillator https://en.wikipedia.org/wiki/GPS_disciplined_oscillator
XMPPwocky 3y ago
> can't exactly use GPS time for synchronization if you're measuring GPS interference Can other GNSSes (Galileo/BeiDou/GLONASS/etc) give usable timestamps? Seems like it'd be tricky for a jammer to target all of them simultaneously. (Of course, since they'd be on a different band, unless your SDR is wideband enough you'd need two RX heads which gives you potential issues with phase drift between the tuning VCOs even if your sampling is coherent). Perhaps a sufficiently directional antenna/phased array (for getting an actual satellite signal) as well as an omnidirectional one (for picking up the jamming signal) could get you somewhere... Or perhaps one could look at computing AoA at each receiver site (using MIMO-y techniques, e.g. Kraken/KerberosSDR) and triangulating based on angles instead, which wouldn't require synchronizing physically-distant sites at all... The problem definitely seems soluble, though I don't have the technical background to know how realistic that is.
- mNovak 3y ago
  >> Can other GNSS give usable timestamps? Seems like it'd be tricky for a jammer to target all of them Actually the opposite; GNSS systems are all purposely designed to operate at virtually the same frequency (check out this figure [1]) while cleverly not interfering with each other. There are sub-bands within each constellation too (L1,L2,L5 etc) but it's very easy to pump out wideband noise across all the GNSS bands. [1] https://www.researchgate.net/figure/The-spectrum-of-current-GNSS-signals-in-15-17-GHz_fig1_259435730 https://www.researchgate.net/figure/The-spectrum-of-current-...

by mNovak 3y ago

toomuchtodo 3y ago

Very helpful (as well as sibling comment by pierat). Thank you both. I will need to do some research with regards to pairing SDRs with local disciplined clock that can tolerate temporary loss of remote (stratum-0,1) time precision. https://en.wikipedia.org/wiki/GPS_disciplined_oscillator https://en.wikipedia.org/wiki/GPS_disciplined_oscillator

XMPPwocky 3y ago

> can't exactly use GPS time for synchronization if you're measuring GPS interference Can other GNSSes (Galileo/BeiDou/GLONASS/etc) give usable timestamps? Seems like it'd be tricky for a jammer to target all of them simultaneously. (Of course, since they'd be on a different band, unless your SDR is wideband enough you'd need two RX heads which gives you potential issues with phase drift between the tuning VCOs even if your sampling is coherent). Perhaps a sufficiently directional antenna/phased array (for getting an actual satellite signal) as well as an omnidirectional one (for picking up the jamming signal) could get you somewhere... Or perhaps one could look at computing AoA at each receiver site (using MIMO-y techniques, e.g. Kraken/KerberosSDR) and triangulating based on angles instead, which wouldn't require synchronizing physically-distant sites at all... The problem definitely seems soluble, though I don't have the technical background to know how realistic that is.

mNovak 3y ago
>> Can other GNSS give usable timestamps? Seems like it'd be tricky for a jammer to target all of them Actually the opposite; GNSS systems are all purposely designed to operate at virtually the same frequency (check out this figure [1]) while cleverly not interfering with each other. There are sub-bands within each constellation too (L1,L2,L5 etc) but it's very easy to pump out wideband noise across all the GNSS bands. [1] https://www.researchgate.net/figure/The-spectrum-of-current-GNSS-signals-in-15-17-GHz_fig1_259435730 https://www.researchgate.net/figure/The-spectrum-of-current-...