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. DC bias in USRP2 & WBX (John Carlson)
2. Re: DC bias in USRP2 & WBX (Matt Ettus)
3. Re: DC bias in USRP2 & WBX (John Carlson)
4. Question about basic daughterboard with UHD (Per Zetterberg)
----------------------------------------------------------------------
Message: 1
Date: Sun, 27 Mar 2011 19:14:38 -0400
From: John Carlson <[email protected]>
To: [email protected]
Subject: [USRP-users] DC bias in USRP2 & WBX
Message-ID:
<[email protected]>
Content-Type: text/plain; charset="iso-8859-1"
I have seen various explanations in other postings about the causes of large
DC bias with the USRP2 (with WBX RF). I am also seeing this problem.
Recently, someone suggested that it was the result of the LO leakage from
the direct down-conversion in the quadrature demodulator. However, the
ADL5387 datasheet explicitly indicates that LO port to RF port leakage is <
-70dBc, which seems much smaller than what we (and probably others) have
observed. An old posting from Matt Ettus suggested that integer truncation
in the receive chain causes a bias which can be corrected by adding back 0.5
to the data. This has not proven to solve the problem. We have also seen
some cases where the bias exists for some tuning frequencies and not in
others. Could the folks at Ettus (or others if you have directly
experienced this issue) please respond with some solid guidance on this
issue? Filtering it out in software and other signal processing workarounds
are not acceptable for our application as it would overly burden our
processors which are already at capacity.
Thanks
John
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
<http://lists.ettus.com/pipermail/usrp-users_lists.ettus.com/attachments/20110327/777bd4fd/attachment-0001.html>
------------------------------
Message: 2
Date: Sun, 27 Mar 2011 16:35:09 -0700
From: Matt Ettus <[email protected]>
To: John Carlson <[email protected]>
Cc: [email protected]
Subject: Re: [USRP-users] DC bias in USRP2 & WBX
Message-ID: <[email protected]>
Content-Type: text/plain; charset=ISO-8859-1
First, you need to clarify what you are talking about. DC bias on RX is
very different from LO->RF leakage. So is your concern the DC bias on
the received digital signal? If so, have you actually measured it?
How? What values have you measured? Under what conditions?
Are you using the UHD drivers? Are you running the transmitter at the
same time? Is it at the same frequency?
The residual DC offset, even without the correction I mentioned
previously, should be below about -90 dBFS.
Matt
On 03/27/2011 04:14 PM, John Carlson wrote:
> I have seen various explanations in other postings about the causes of
> large DC bias with the USRP2 (with WBX RF). I am also seeing this
> problem. Recently, someone suggested that it was the result of the LO
> leakage from the direct down-conversion in the quadrature demodulator.
> However, the ADL5387 datasheet explicitly indicates that LO port to RF
> port leakage is < -70dBc, which seems much smaller than what we (and
> probably others) have observed. An old posting from Matt Ettus
> suggested that integer truncation in the receive chain causes a bias
> which can be corrected by adding back 0.5 to the data. This has not
> proven to solve the problem. We have also seen some cases where the bias
> exists for some tuning frequencies and not in others. Could the folks
> at Ettus (or others if you have directly experienced this issue) please
> respond with some solid guidance on this issue? Filtering it out in
> software and other signal processing workarounds are not acceptable for
> our application as it would overly burden our processors which are
> already at capacity.
> Thanks
> John
>
>
>
> _______________________________________________
> USRP-users mailing list
> [email protected]
> http://lists.ettus.com/mailman/listinfo/usrp-users_lists.ettus.com
------------------------------
Message: 3
Date: Sun, 27 Mar 2011 21:02:13 -0400
From: John Carlson <[email protected]>
To: Matt Ettus <[email protected]>, usrp-users
<[email protected]>
Subject: Re: [USRP-users] DC bias in USRP2 & WBX
Message-ID:
<[email protected]>
Content-Type: text/plain; charset="iso-8859-1"
Hi Matt-
Given your response, it thankfully sounds like I'm doing something wrong. I
can tell you that this is receive only with no transmitter running, and is
not using the UHD drivers. However, we have run some tests with UHD and
have observed similar results. We will run some additional tests and
precisely measure what we see and report it back to the group for discussion
and comment later this week.
Thanks
John
On Sun, Mar 27, 2011 at 7:35 PM, Matt Ettus <[email protected]> wrote:
>
>
> First, you need to clarify what you are talking about. DC bias on RX is
> very different from LO->RF leakage. So is your concern the DC bias on
> the received digital signal? If so, have you actually measured it?
> How? What values have you measured? Under what conditions?
>
> Are you using the UHD drivers? Are you running the transmitter at the
> same time? Is it at the same frequency?
>
> The residual DC offset, even without the correction I mentioned
> previously, should be below about -90 dBFS.
>
> Matt
>
> On 03/27/2011 04:14 PM, John Carlson wrote:
> > I have seen various explanations in other postings about the causes of
> > large DC bias with the USRP2 (with WBX RF). I am also seeing this
> > problem. Recently, someone suggested that it was the result of the LO
> > leakage from the direct down-conversion in the quadrature demodulator.
> > However, the ADL5387 datasheet explicitly indicates that LO port to RF
> > port leakage is < -70dBc, which seems much smaller than what we (and
> > probably others) have observed. An old posting from Matt Ettus
> > suggested that integer truncation in the receive chain causes a bias
> > which can be corrected by adding back 0.5 to the data. This has not
> > proven to solve the problem. We have also seen some cases where the bias
> > exists for some tuning frequencies and not in others. Could the folks
> > at Ettus (or others if you have directly experienced this issue) please
> > respond with some solid guidance on this issue? Filtering it out in
> > software and other signal processing workarounds are not acceptable for
> > our application as it would overly burden our processors which are
> > already at capacity.
> > Thanks
> > John
> >
> >
> >
> > _______________________________________________
> > USRP-users mailing list
> > [email protected]
> > http://lists.ettus.com/mailman/listinfo/usrp-users_lists.ettus.com
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
<http://lists.ettus.com/pipermail/usrp-users_lists.ettus.com/attachments/20110327/c9f31eb6/attachment-0001.html>
------------------------------
Message: 4
Date: Mon, 28 Mar 2011 09:25:24 +0000
From: Per Zetterberg <[email protected]>
To: "[email protected]" <[email protected]>
Subject: [USRP-users] Question about basic daughterboard with UHD
Message-ID: <[email protected]>
Content-Type: text/plain; charset="us-ascii"
Hi Josh and List,
Can I use the two SMA inputs as the I and Q channels of a direct downconversion
receiver ?
MVH/
Per
------------------------------
_______________________________________________
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 48
*****************************************
