krakenSDR

686 views
Skip to first unread message

Hamish Barker

unread,
Oct 15, 2021, 10:42:03 PM10/15/21
to Society of Amateur Radio Astronomers

I am not sure if the coherence of the receivers of this unit will really be enough for interferometry, but am willing to risk $300 to find out.  ( I also have a LimeSDR on order but this one has five channels. If it works, 5 baselines it could be really something.


Cheers,
Hamish

Marcus D. Leech

unread,
Oct 15, 2021, 10:46:07 PM10/15/21
to sara...@googlegroups.com
They just use 5 RTL-SDRs, with a common clock, and a switched noise generator to allow timing and phase alignment.

What isn't clear is how well they've "taken care of" all that inside their drivers.  If not, then the application has to take care of it at start-up, and
  the computations aren't trivial (doable, mind, just a pain).

With 5 mutually-coherent inputs, and non-redundant antenna spacings, you get 10 baselines with 5 inputs.

5*(5-1) / 2



--
--
You received this message because you are subscribed to the Google
Groups "Society of Amateur Radio Astronomers" group.
To post to this group, send email to sara...@googlegroups.com
To unsubscribe from this group, send email to
sara-list-...@googlegroups.com
For more options, visit this group at
http://groups.google.com/group/sara-list?hl=en
---
You received this message because you are subscribed to the Google Groups "Society of Amateur Radio Astronomers" group.
To unsubscribe from this group and stop receiving emails from it, send an email to sara-list+...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/sara-list/c229abb1-eac1-499f-8788-76121d7ffc09n%40googlegroups.com.

Marcus D. Leech

unread,
Oct 15, 2021, 11:46:53 PM10/15/21
to sara...@googlegroups.com
On 2021-10-15 10:42 p.m., Hamish Barker wrote:
>
I've just had a chat with Carl Laufer, one of the project management
people for KrakenSDR.  Their vision is that their "DAQ" layer can be
used to talk to the
  radios as a "coherent bundle"--so that layer can take care of all of
the ugly details of creating and maintaining that "coherent bundle". 
That DAQ layer can
  either run on your host PC, with the KrakenSDR plugged in directly,
OR you can have an rPi4 "in the middle" providing that service, and you
get that
  "coherent bundle" delivered to you over 1GiGe.

I've set some wheels in motion to see if some of the Gnu Radio
developers might be interested in making an interface to this "DAQ" layer.

The bandwidths of this thing are *tiny* in comparison to what the "big
boys" use for their coherent arrays.  But it does open up some very
  interesting possibilities.

With 5 antennas, you can do some not-entirely-crude synthesis imaging.



Larry Mayfield

unread,
Oct 16, 2021, 10:53:16 AM10/16/21
to sara...@googlegroups.com

I think I will start watching Banggood because I suspect that their sources are already working to duplicate this…   Iwol dbe. Looks very nifty.

 

Larry

Pahrump, NV

Anthony

unread,
Oct 16, 2021, 11:07:07 AM10/16/21
to sara...@googlegroups.com
Just watch the reviews on the quality of the parts. I've been reviewing Banggood, just read closely the reviews and make your own purchase decision,  based on the validity of the persons providing the reviews.

Anthony

unread,
Oct 16, 2021, 1:06:12 PM10/16/21
to sara...@googlegroups.com
What are your thoughts on the KerberosSDR 4x Coherent RTL-SDR with calibration board for synchronizing samples with the noise source?

Seems to be available. There is an open-source DSP code base mentioned, does that mean GNU can run on this but requires a good skillset with GNU & Python? the website only mentions direction-finding & 2-RX passive radar. No mention of beamforming and RA interferometry, which I'm unsure if this would work for an interferometer. Guess that would depend on the custom software used, and the calibration board for synchronizing the noise samples obtained from two antenna receivers?

The ettus B210 is 2-Rx, 2-Tx, and runs with GNU radio, but I saw on a forum the learning curve is steep. The reviewer used some real choice words, I'm being mild in regurgitating his review.  Compared to eBays foreign countries, offering the USRP B210 board is cheaper but may be of lower quality. Guess any work with an interferometer is a steep learning curve. 

Ettus B210 2x2 (board only), very pricey for only 2, Rx but it comes with 2, Tx. and uses the AD9361 to provide coherent mimo capabilities. 
Signal processing, the B210 is using the Spartan XC6SLX150-2FGG484I FPGA chip. 

Ettus USRP B210  (at Amazon, same price but aluminum enclosure & 6V power supply included), which for some mysterious reasons is no longer available, from last week?

A few others 4, Rx out there, but I'm always apprehensive to purchase from overseas

bladeRF 2.0 micro xA4 Software Defined Radio


BladeRF 2.0 xA4kit (pricey)

This is extremely tedious to review all of these offerings and looks like the best candidate is to purchase the KrakenSDR, (for future growth planning) and wait till the end of March 2022 (if you purchase now) for shipment. Or fork out  $1500 + $100 (enclosure), not counting the power supply cost, on the B210. One reviewer said, after weeks of working on the Ettus B210, he still could not get it to work. But I'm unsure of his skill set with GNU radio. 
Would the B210 work with Virgo, since Virgo use GNU & Python?



--
--
You received this message because you are subscribed to the Google
Groups "Society of Amateur Radio Astronomers" group.
To post to this group, send email to sara...@googlegroups.com
To unsubscribe from this group, send email to
sara-list-...@googlegroups.com
For more options, visit this group at
http://groups.google.com/group/sara-list?hl=en
---
You received this message because you are subscribed to the Google Groups "Society of Amateur Radio Astronomers" group.
To unsubscribe from this group and stop receiving emails from it, send an email to sara-list+...@googlegroups.com.

Marcus D. Leech

unread,
Oct 16, 2021, 1:28:02 PM10/16/21
to sara...@googlegroups.com
On 2021-10-16 1:05 p.m., Anthony wrote:
What are your thoughts on the KerberosSDR 4x Coherent RTL-SDR with calibration board for synchronizing samples with the noise source?

Seems to be available. There is an open-source DSP code base mentioned, does that mean GNU can run on this but requires a good skillset with GNU & Python? the website only mentions direction-finding & 2-RX passive radar. No mention of beamforming and RA interferometry, which I'm unsure if this would work for an interferometer. Guess that would depend on the custom software used, and the calibration board for synchronizing the noise samples obtained from two antenna receivers?
The KrakenSDR is the "next gen" KerberosSDR--done by the same people.


The ettus B210 is 2-Rx, 2-Tx, and runs with GNU radio, but I saw on a forum the learning curve is steep. The reviewer used some real choice words, I'm being mild in regurgitating his review.  Compared to eBays foreign countries, offering the USRP B210 board is cheaper but may be of lower quality. Guess any work with an interferometer is a steep learning curve.
The "learning curve" for ANY SDR is steep unless you're just going to use a pre-canned application.


Ettus B210 2x2 (board only), very pricey for only 2, Rx but it comes with 2, Tx. and uses the AD9361 to provide coherent mimo capabilities. 
Signal processing, the B210 is using the Spartan XC6SLX150-2FGG484I FPGA chip. 

Ettus USRP B210  (at Amazon, same price but aluminum enclosure & 6V power supply included), which for some mysterious reasons is no longer available, from last week?

A few others 4, Rx out there, but I'm always apprehensive to purchase from overseas

bladeRF 2.0 micro xA4 Software Defined Radio


BladeRF 2.0 xA4kit (pricey)

This is extremely tedious to review all of these offerings and looks like the best candidate is to purchase the KrakenSDR, (for future growth planning) and wait till the end of March 2022 (if you purchase now) for shipment. Or fork out  $1500 + $100 (enclosure), not counting the power supply cost, on the B210. One reviewer said, after weeks of working on the Ettus B210, he still could not get it to work. But I'm unsure of his skill set with GNU radio. 
Would the B210 work with Virgo, since Virgo use GNU & Python?
The B210 should work just fine with Virgo, as far as I know--it uses one of the two generic hardware interfaces (SoapySDR or gr-osmosdr).

There are literally *thousands* of USRP B210 in the field running dozens and dozens of different applications.  If someone "cannot get it to work", then they just
  don't have the correct skillset.  If you buy an SDR expecting it to be just like an olde-skool hardware (analog) radio, then you'll be massively disappointed.
  They are simply *components* in an overall radio *system*.  If what you want to do is "casual airwave surfing", then generic applications like GQRX, SDRSharp,
  SDRAngel, etc, will generally satisfy you.  If you have something *else* in mind, then unless your particular application has already been written and packaged,
  the SDR hardware will be of very little help on its own, and you'll need to learn some programming and DSP skills. 

It's a bit like buying an automotive water-pump, alternator, and a few other things, laying them expectantly on a table, and hoping for a friendly blue "Build thyself into a car"
  button to suddenly appear.  It's not going to happen.

There are LOTs of people who buy SDRs, expecting that they'll by some wizardry currently unknown to humankind, be already "capable" of doing whatever it is they
  want to do with a radio.  If one considers a couple of interesting sets:

  (A) "Interesting things one might do with radio technology"
  (B) "Interesting and useful things one might do with a computer"

It becomes clear very quickly that those sets are both very, very, very large.  Which means that no finite effort is likely to produce every possible thing that is in
  the intersection of those two sets.

The same sort of thing happened with early personal computers.  People bought them thinking they'd already know how to do "all the things", because that's
  how they were marketed.  Only to discover that no, your computer was not "born" knowing how to translate from German into Algebra.  Some programming may
  be required.





Anthony

unread,
Oct 16, 2021, 1:55:19 PM10/16/21
to sara...@googlegroups.com
Thanks Marcus,

Makes a whole lot of sense. I'll stay with the Ettus B210, using the BAA & Virgo with GNU. I have some Python skills but never used them for RA. I'll develop my skill set further. 

To view this discussion on the web visit https://groups.google.com/d/msgid/sara-list/7dea6a11-7a99-dddf-6765-770347449dd1%40gmail.com.

Hamish Barker

unread,
Oct 16, 2021, 6:18:37 PM10/16/21
to sara...@googlegroups.com
Thanks for inquiring and the update Marcus!

Apparently some of their team is in Auckland, not so far from me. 

Indeed, if it works and signal to noise is acceptable, 10 non redundant baselines could be really something. A Mini chime trough could be a cheap way to large aperture drift line scan, or 5 tracking dishes on an e/w orientation for imaging.

--
--
You received this message because you are subscribed to the Google
Groups "Society of Amateur Radio Astronomers" group.
To post to this group, send email to sara...@googlegroups.com
To unsubscribe from this group, send email to
sara-list-...@googlegroups.com
For more options, visit this group at
http://groups.google.com/group/sara-list?hl=en
---
You received this message because you are subscribed to a topic in the Google Groups "Society of Amateur Radio Astronomers" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/sara-list/ZFJEfZeGEPY/unsubscribe.
To unsubscribe from this group and all its topics, send an email to sara-list+...@googlegroups.com.

bsn...@gmail.com

unread,
Oct 16, 2021, 9:14:55 PM10/16/21
to Society of Amateur Radio Astronomers
Words of wisdom: 
(A) "Interesting things one might do with radio technology"
  (B) "Interesting and useful things one might do with a computer"

Marcus, as you well know, either one of these sets of "interesting things" can keep you amused, gainfully employed, or otherwise occupied for a lifetime.
Built first radio in 1958, first programming course FORTAN IV 1969...still learning and playing with both thanks to amateur radio astronomy.

Marcus D. Leech

unread,
Oct 16, 2021, 9:31:05 PM10/16/21
to sara...@googlegroups.com
On 2021-10-16 9:14 p.m., bsn...@gmail.com wrote:
Words of wisdom: 
(A) "Interesting things one might do with radio technology"
  (B) "Interesting and useful things one might do with a computer"

Marcus, as you well know, either one of these sets of "interesting things" can keep you amused, gainfully employed, or otherwise occupied for a lifetime.
Built first radio in 1958, first programming course FORTAN IV 1969...still learning and playing with both thanks to amateur radio astronomy.
Indeed.  Which is why I'm always amused when folks arrive at some newfangled technical discipline like SDR and ask "why doesn't it already know how to do
  <thing> surely EVERYONE is interested in <thing> and that would be the most important thing to implement."

I built my first (crystal) radio with my dad around 1970 or so.  Started getting into hobby electronics around 1973 when a school chum of mine was the benefactor
  of access to the electronics lab at the Erindale campus of University of Toronto--his father was Dean at the time...

Started messing-about with computers and software in 1977 or so, and then got a part-time job at the same university in computing in 1979.  I STILL haven't built
 all the things I could ever imagine :) :)


Paul Oxley

unread,
Oct 16, 2021, 10:54:02 PM10/16/21
to sara...@googlegroups.com
cMarcus

Your dates are questionable or wrong.

I personally listened to the AM radio long before 1958. It included stations in the U.S. and Mexico. I don't remember Canada. It was probably too close to Iowa for night time skip. I remember Del Rio, TX and WLS in Chicago. We had an FM receiver before any stations were broadcasting.

I watched TV in 1952 (the Presidential innoguration). The school assigned everyone to a house that had a TV. We did not. FM was tied to TV and occupied channel 1 of the TV allocation.

Paul

Marcus D. Leech

unread,
Oct 16, 2021, 11:43:04 PM10/16/21
to sara...@googlegroups.com
On 2021-10-16 10:53 p.m., Paul Oxley wrote:
cMarcus

Your dates are questionable or wrong.

I personally listened to the AM radio long before 1958. It included stations in the U.S. and Mexico. I don't remember Canada. It was probably too close to Iowa for night time skip. I remember Del Rio, TX and WLS in Chicago. We had an FM receiver before any stations were broadcasting.

I watched TV in 1952 (the Presidential innoguration). The school assigned everyone to a house that had a TV. We did not. FM was tied to TV and occupied channel 1 of the TV allocation.

Paul
??????


James Brown

unread,
Oct 17, 2021, 1:24:47 AM10/17/21
to Society of Amateur Radio Astronomers
BTW is was the station engineer for an AM station (KLO Ogden Utah) in 1958 
... so I win.
Dilbert_and_Old_Guys.png

Hamish Barker

unread,
Sep 5, 2022, 6:38:31 PM9/5/22
to Society of Amateur Radio Astronomers
https://github.com/krakenrf/krakensdr_docs/wiki/09.-GNU-Radio-Block

the kraken GNU radio block looks potentially quite useable for amateur radioastronomy interferometry. the documentation of krakensdr makes clear that they see its use for this.

5 receivers could potentially make 10 non redundant baselines. but even one or three pairs of coherent receivers will be awesome if it works. 

Anthony

unread,
Sep 5, 2022, 6:49:10 PM9/5/22
to Society of Amateur Radio Astronomers
I have not had a chance to test out the GNU Radio block yet, with KrakenSDR. 

Hoping to start by months end.

--
--
You received this message because you are subscribed to the Google
Groups "Society of Amateur Radio Astronomers" group.
To post to this group, send email to sara...@googlegroups.com
To unsubscribe from this group, send email to
sara-list-...@googlegroups.com
For more options, visit this group at
http://groups.google.com/group/sara-list?hl=en
---
You received this message because you are subscribed to the Google Groups "Society of Amateur Radio Astronomers" group.
To unsubscribe from this group and stop receiving emails from it, send an email to sara-list+...@googlegroups.com.

Anthony

unread,
Sep 5, 2022, 7:29:19 PM9/5/22
to Society of Amateur Radio Astronomers
The Heimdall DAQ Server software looks like a bitof and installs, some config.ini file tweaks. Runs on either Pi4, x86 or ARM computer, requiring 4 cores (tested on a Pi). The minimum looks like 2 cores for a PC.

Hamish Barker

unread,
Sep 5, 2022, 7:49:28 PM9/5/22
to sara...@googlegroups.com
good luck, keep us informed! 

To view this discussion on the web visit https://groups.google.com/d/msgid/sara-list/CAKU65TyFe85KbCAYFg28K3zPJSZkS%3DTkzNPp1z-8y6ZBXPeGxw%40mail.gmail.com.

Marcus D. Leech

unread,
Sep 5, 2022, 8:05:42 PM9/5/22
to sara...@googlegroups.com
On 2022-09-05 19:49, Hamish Barker wrote:
good luck, keep us informed!
The only limitation with this kit, assuming it really is long-term coherent, is that it's obviously limited to the bandwidth
  available with the RTL-SDR, which is no greater than 2.5MHz.

Good for "stamp collecting" the pantheon of the "usual bright sources", albeit, with greater detail than you could do
  with a single dish...


Anthony

unread,
Sep 5, 2022, 8:24:17 PM9/5/22
to Society of Amateur Radio Astronomers
Will do, thanks Hamish!

Anthony

unread,
Sep 8, 2022, 6:44:26 PM9/8/22
to sara...@googlegroups.com, hamish...@gmail.com
Hi Hamish,

I decided to move forward and download the installs for the KrakenSDR. My first attempt was a manual install using an X86 machine, instead of a PI4. 
There were a few glitches along the way, like the main KrakenSDR Radar finder main page didn't pull up, even after going into the krakensdr_daq

Reply all
Reply to author
Forward
0 new messages