Send USRP-users mailing list submissions to
[email protected]
To subscribe or unsubscribe via the World Wide Web, visit
http://lists.ettus.com/mailman/listinfo/usrp-users_lists.ettus.com
or, via email, send a message with subject or body 'help' to
[email protected]
You can reach the person managing the list at
[email protected]
When replying, please edit your Subject line so it is more specific
than "Re: Contents of USRP-users digest..."
Today's Topics:
1. Re: Timing and Frequency Synchronisation in MATLAB (Nick Foster)
2. Re: UHD transport protocol (Josh Blum)
3. Re: Timing and Frequency Synchronisation in MATLAB (Mike McLernon)
4. Re: Timing and Frequency Synchronisation in MATLAB (Nick Foster)
5. Re: Question for "USRP2 + RFX2400" (Josh Blum)
6. Turning off the digital down converter(DDC) in USRP2
(Song, Mujun CAPT KR USAF AETC AFIT/ENG)
----------------------------------------------------------------------
Message: 1
Date: Tue, 08 Mar 2011 10:18:33 -0800
From: Nick Foster <[email protected]>
To: [email protected]
Cc: [email protected]
Subject: Re: [USRP-users] Timing and Frequency Synchronisation in
MATLAB
Message-ID: <1299608313.3438.11.camel@crapshoot>
Content-Type: text/plain; charset="UTF-8"
On Tue, 2011-03-08 at 12:23 +0200, Ryan van den Bergh wrote:
> Hi Everyone,
>
> Firstly thanks for everyone who has helped me in the past with my USRP2
> and Simulink issues - I really appreciate it!!
>
> At the moment I need to create a full transceiver system in Simulink
> with the USRP2s and I need timing and frequency synchronisation for my
> frames.
>
> Now before I go any further, I have read the responses from Mike
> McLernon at Mathworks regarding Barker codes, cross-correlation and
> pulse shaping (Many thanks for those e-mails Mike - they helped
> immensely!!!), but what I'm actually wondering is whether the reference
> clock and pps could be used to provide a sort of preliminary
> synchronisation between the transmitting and receiving USRP2s. All the
> literature I have found thus far relates only seems to apply to GNURadio
> systems.
>
> My thought process (and please feel free to correct me if I'm completely
> wrong) is that by attaching two function generators to the USRP2s (one
> to both reference clock ports, and one to both pps ports) I will be able
> to largely eliminate frequency drift and timing offset. I believe this
> would make the barker code preambles and cross-correlation far more
> accurate when determining the start of the frame and greatly assist in
> the case of frames that contain OFDM symbols.
>
> So to summarise, I'll list my questions:
>
> 1. Would the addition of signals on the ref clock and pps ports assist
> in frame synchronisation?
Sure. It's not a very realistic simulation of a real digital radio, but
if you don't need that, then it'll save you some grief. That said, it
won't do anything for your frame synchronization, just for your
frequency synchronization. Both are necessary in a real digital radio
system. You will still have to do some sort of framing on the transmit
side, and preamble detection on the receive side, to detect the start of
your frames. This is pretty straightforward, and I'm pretty sure
Simulink blocks exist to just "make it work".
> 2. Would the use of a standard function generator be
> appropriate/applicable or would I need more specialised hardware?
That's fine. Give it 10MHz at +10dBm and it'll sync.
> 3. Depending on the answer above, can I use two separate function
> generators with the appropriate signal characteristics?
Not necessary.
> 4. Bearing in mind that I am using MATLAB/Simulink which requires old
> versions of the USRP2 firmware, do I have to do anything to the USRP2s
> to get them to use these signals?
Here's the problem. I don't think that the current Simulink driver
supports locking to the ref clock input. I'm happy to be corrected, but
you might have to 1) use Gnuradio and either gr-uhd or the older raw
Ethernet driver, or 2) wait for Mathworks to release a UHD-compatible
Simulink driver (any day now?).
--n
>
> Kind regards,
>
> Ryan
>
> _______________________________________________
> USRP-users mailing list
> [email protected]
> http://lists.ettus.com/mailman/listinfo/usrp-users_lists.ettus.com
------------------------------
Message: 2
Date: Tue, 08 Mar 2011 10:41:47 -0800
From: Josh Blum <[email protected]>
To: [email protected]
Subject: Re: [USRP-users] UHD transport protocol
Message-ID: <[email protected]>
Content-Type: text/plain; charset=ISO-8859-1
On 03/08/2011 05:57 AM, Scott Johnston wrote:
>
> Hello,
>
> Can someone tell me how the UHD is using udp and vrt? The documentation
> http://www.ettus.com/uhd_docs/manual/html/transport.html just mentions
> udp, but in the uhd/host/lib/usrp/usrp2 io_impl.cpp file it calls the
> vrt_packet_handler.
>
> Thanks for any info
>
I guess you could say that VRT is at the application layer in the OSI
model. -Josh
> Scott
>
> _______________________________________________
> USRP-users mailing list
> [email protected]
> http://lists.ettus.com/mailman/listinfo/usrp-users_lists.ettus.com
------------------------------
Message: 3
Date: Tue, 8 Mar 2011 18:48:11 +0000
From: Mike McLernon <[email protected]>
To: "[email protected]" <[email protected]>
Cc: Christoph Georgi <[email protected]>,
"[email protected]" <[email protected]>
Subject: Re: [USRP-users] Timing and Frequency Synchronisation in
MATLAB
Message-ID:
<e3879be9a282cb45aab7ce258a9ae48f06f...@exmb-01-ah.ad.mathworks.com>
Content-Type: text/plain; charset="utf-8"
Hi Ryan,
The UDP-based Simulink I/O blocks should not be sensitive to the source of the
clock, whether it is internally generated or externally input. We have not
tested this capability, however, so I cannot guarantee that it will work. So,
a hardware question to the group -- if an external clock is connected to the
board, does the UDP firmware automatically lock to it?
Mike
-----Original Message-----
From: Nick Foster [mailto:[email protected]]
Sent: Tuesday, March 08, 2011 1:19 PM
To: [email protected]
Cc: [email protected]; Mike McLernon
Subject: Re: [USRP-users] Timing and Frequency Synchronisation in MATLAB
On Tue, 2011-03-08 at 12:23 +0200, Ryan van den Bergh wrote:
> Hi Everyone,
>
> Firstly thanks for everyone who has helped me in the past with my USRP2
> and Simulink issues - I really appreciate it!!
>
> At the moment I need to create a full transceiver system in Simulink
> with the USRP2s and I need timing and frequency synchronisation for my
> frames.
>
> Now before I go any further, I have read the responses from Mike
> McLernon at Mathworks regarding Barker codes, cross-correlation and
> pulse shaping (Many thanks for those e-mails Mike - they helped
> immensely!!!), but what I'm actually wondering is whether the reference
> clock and pps could be used to provide a sort of preliminary
> synchronisation between the transmitting and receiving USRP2s. All the
> literature I have found thus far relates only seems to apply to GNURadio
> systems.
>
> My thought process (and please feel free to correct me if I'm completely
> wrong) is that by attaching two function generators to the USRP2s (one
> to both reference clock ports, and one to both pps ports) I will be able
> to largely eliminate frequency drift and timing offset. I believe this
> would make the barker code preambles and cross-correlation far more
> accurate when determining the start of the frame and greatly assist in
> the case of frames that contain OFDM symbols.
>
> So to summarise, I'll list my questions:
>
> 1. Would the addition of signals on the ref clock and pps ports assist
> in frame synchronisation?
Sure. It's not a very realistic simulation of a real digital radio, but
if you don't need that, then it'll save you some grief. That said, it
won't do anything for your frame synchronization, just for your
frequency synchronization. Both are necessary in a real digital radio
system. You will still have to do some sort of framing on the transmit
side, and preamble detection on the receive side, to detect the start of
your frames. This is pretty straightforward, and I'm pretty sure
Simulink blocks exist to just "make it work".
> 2. Would the use of a standard function generator be
> appropriate/applicable or would I need more specialised hardware?
That's fine. Give it 10MHz at +10dBm and it'll sync.
> 3. Depending on the answer above, can I use two separate function
> generators with the appropriate signal characteristics?
Not necessary.
> 4. Bearing in mind that I am using MATLAB/Simulink which requires old
> versions of the USRP2 firmware, do I have to do anything to the USRP2s
> to get them to use these signals?
Here's the problem. I don't think that the current Simulink driver
supports locking to the ref clock input. I'm happy to be corrected, but
you might have to 1) use Gnuradio and either gr-uhd or the older raw
Ethernet driver, or 2) wait for Mathworks to release a UHD-compatible
Simulink driver (any day now?).
--n
>
> Kind regards,
>
> Ryan
>
> _______________________________________________
> USRP-users mailing list
> [email protected]
> http://lists.ettus.com/mailman/listinfo/usrp-users_lists.ettus.com
------------------------------
Message: 4
Date: Tue, 08 Mar 2011 10:54:20 -0800
From: Nick Foster <[email protected]>
To: Mike McLernon <[email protected]>
Cc: Christoph Georgi <[email protected]>,
"[email protected]" <[email protected]>,
"[email protected]" <[email protected]>
Subject: Re: [USRP-users] Timing and Frequency Synchronisation in
MATLAB
Message-ID: <1299610460.3438.12.camel@crapshoot>
Content-Type: text/plain; charset="UTF-8"
On Tue, 2011-03-08 at 18:48 +0000, Mike McLernon wrote:
> Hi Ryan,
>
> The UDP-based Simulink I/O blocks should not be sensitive to the source of
> the clock, whether it is internally generated or externally input. We have
> not tested this capability, however, so I cannot guarantee that it will work.
> So, a hardware question to the group -- if an external clock is connected to
> the board, does the UDP firmware automatically lock to it?
>
No. The default is free-running unless commanded to sync to the external
reference.
--n
> Mike
>
>
> -----Original Message-----
> From: Nick Foster [mailto:[email protected]]
> Sent: Tuesday, March 08, 2011 1:19 PM
> To: [email protected]
> Cc: [email protected]; Mike McLernon
> Subject: Re: [USRP-users] Timing and Frequency Synchronisation in MATLAB
>
> On Tue, 2011-03-08 at 12:23 +0200, Ryan van den Bergh wrote:
> > Hi Everyone,
> >
> > Firstly thanks for everyone who has helped me in the past with my USRP2
> > and Simulink issues - I really appreciate it!!
> >
> > At the moment I need to create a full transceiver system in Simulink
> > with the USRP2s and I need timing and frequency synchronisation for my
> > frames.
> >
> > Now before I go any further, I have read the responses from Mike
> > McLernon at Mathworks regarding Barker codes, cross-correlation and
> > pulse shaping (Many thanks for those e-mails Mike - they helped
> > immensely!!!), but what I'm actually wondering is whether the reference
> > clock and pps could be used to provide a sort of preliminary
> > synchronisation between the transmitting and receiving USRP2s. All the
> > literature I have found thus far relates only seems to apply to GNURadio
> > systems.
> >
> > My thought process (and please feel free to correct me if I'm completely
> > wrong) is that by attaching two function generators to the USRP2s (one
> > to both reference clock ports, and one to both pps ports) I will be able
> > to largely eliminate frequency drift and timing offset. I believe this
> > would make the barker code preambles and cross-correlation far more
> > accurate when determining the start of the frame and greatly assist in
> > the case of frames that contain OFDM symbols.
> >
> > So to summarise, I'll list my questions:
> >
> > 1. Would the addition of signals on the ref clock and pps ports assist
> > in frame synchronisation?
>
> Sure. It's not a very realistic simulation of a real digital radio, but
> if you don't need that, then it'll save you some grief. That said, it
> won't do anything for your frame synchronization, just for your
> frequency synchronization. Both are necessary in a real digital radio
> system. You will still have to do some sort of framing on the transmit
> side, and preamble detection on the receive side, to detect the start of
> your frames. This is pretty straightforward, and I'm pretty sure
> Simulink blocks exist to just "make it work".
>
> > 2. Would the use of a standard function generator be
> > appropriate/applicable or would I need more specialised hardware?
>
> That's fine. Give it 10MHz at +10dBm and it'll sync.
>
> > 3. Depending on the answer above, can I use two separate function
> > generators with the appropriate signal characteristics?
>
> Not necessary.
>
> > 4. Bearing in mind that I am using MATLAB/Simulink which requires old
> > versions of the USRP2 firmware, do I have to do anything to the USRP2s
> > to get them to use these signals?
>
> Here's the problem. I don't think that the current Simulink driver
> supports locking to the ref clock input. I'm happy to be corrected, but
> you might have to 1) use Gnuradio and either gr-uhd or the older raw
> Ethernet driver, or 2) wait for Mathworks to release a UHD-compatible
> Simulink driver (any day now?).
>
> --n
>
> >
> > Kind regards,
> >
> > Ryan
> >
> > _______________________________________________
> > USRP-users mailing list
> > [email protected]
> > http://lists.ettus.com/mailman/listinfo/usrp-users_lists.ettus.com
>
>
------------------------------
Message: 5
Date: Tue, 08 Mar 2011 12:14:05 -0800
From: Josh Blum <[email protected]>
To: [email protected]
Subject: Re: [USRP-users] Question for "USRP2 + RFX2400"
Message-ID: <[email protected]>
Content-Type: text/plain; charset=UTF-8
On 03/08/2011 05:28 AM, ??? wrote:
>
>
> Hello.
>
>
>
> I am trying to implement RF relay system with USRP2.
>
> It?s not so long time to start this.
>
> So I am beginner and have many questions about USRP2.
>
>
>
> My questions are as follows.
>
>
>
> 1. I am using the daughter board ?RFX2400?.
>
> I know that RFX2400 is transceiver and its RX and TX independently operate.
>
> But my result of test is different from the prior knowledge.
>
> Is it possible?
>
>
yes
You will need to set the rx antenna to RX2.
http://www.ettus.com/uhd_docs/doxygen/html/classuhd_1_1usrp_1_1multi__usrp.html#a72b7947cb0c434b98e9915f91b8f8fe0
>
> ????_a.jpg
>
>
>
> I generate the signal from the host and transmit the signal to USRP2 through
> the Ethernet.
>
> Can this signal transmitted from the TX antenna be received to RX antenna?
>
>
>
Yes
>
>
> 2. And I observed the discontinuity of transmitting the signal.
>
> What does the discontinuity in the below picture cause? USRP2 or My Computer?
>
>
>
> ????_b.jpg
>
>
>
> Can a signal be continuously transmitted?
>
>
Yes, but perhaps you are getting underflows when transmitting. When you
try lower sample rates, do you continue to observe underflows?
>
> 3. The last question.
>
> I generated "square signal( Freq=500kHz, samp_rate=4MHz)" using GRC.
>
> I changed USRP2's Frequency(RF singnal's center frequency ? F_c)
>
> (1) F_c=2.3GHz
>
> ????_1.jpg
>
> (2) F_c=2.5GHz
>
> ????_2.jpg
>
> (3) F_c=2.7GHz
>
> ????_3.jpg
>
> I observed square signals generated from the different center frequency
> through the spectrum analyzer.
>
> I found that the unexpected signal from 2.4GHz~2.5GHz band always arise.
>
> Why does this situation happen?
>
>
I believe that the "unexpected signal" is the local oscillator on the
daughterboard. You can move the LO relative to your signal's center
frequency by tuning with an LO offset:
http://www.ettus.com/uhd_docs/doxygen/html/classuhd_1_1usrp_1_1multi__usrp.html#ae7a674a9d012a78dd4f2ded478839124
http://www.ettus.com/uhd_docs/doxygen/html/structuhd_1_1tune__request__t.html#af9d2c5fb89c10024b1acae43e88ebe7f
_josh
------------------------------
Message: 6
Date: Tue, 8 Mar 2011 17:06:50 -0500
From: "Song, Mujun CAPT KR USAF AETC AFIT/ENG"
<[email protected]>
To: <[email protected]>
Subject: [USRP-users] Turning off the digital down converter(DDC) in
USRP2
Message-ID:
<[email protected]>
Content-Type: text/plain; charset="us-ascii"
Hi, all.
I'm working with USRP2 and Simulink to analyze the Spectral Correlation
Function of transmitter signal. I have some problem which comes from
Digital Down Converter(DDC) of USRP2. Because the SCF analysis is based
on RF signal or at least IF signal which show positive and negative
frequency peaks, the digital down conversion made by USRP2 makes the SCF
analysis impossible. As I know, the DDC is implemented in FPGA. Is there
any way to bypass the DDC process in USRP2 to get a IF or RF signal? Or
is there anybody who know how to get IF or RF signal using different
way? And I have to keep the link between USRP2 and Simulink.
Hope to get response from you.
Thanks,
Mujun
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
<http://lists.ettus.com/pipermail/usrp-users_lists.ettus.com/attachments/20110308/a01e9772/attachment.html>
------------------------------
_______________________________________________
USRP-users mailing list
[email protected]
http://lists.ettus.com/mailman/listinfo/usrp-users_lists.ettus.com
End of USRP-users Digest, Vol 7, Issue 14
*****************************************
