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: Receive IQ samples with USRP2 and WBX RF Board (Nick Foster)
2. Re: Data length of USRP2 receiver in Matlab simulink (yqluo)
3. Re: UHD mirror stuck (Jason Abele)
4. Re: Welcome to the "USRP-users" mailing list (zhe yang)
5. Re: Welcome to the "USRP-users" mailing list (Mike McLernon)
----------------------------------------------------------------------
Message: 1
Date: Wed, 09 Mar 2011 10:37:02 -0800
From: Nick Foster <[email protected]>
To: Haenggi Stefan <[email protected]>
Cc: "[email protected]" <[email protected]>
Subject: Re: [USRP-users] Receive IQ samples with USRP2 and WBX RF
Board
Message-ID: <1299695822.24655.33.camel@crapshoot>
Content-Type: text/plain; charset="UTF-8"
On Wed, 2011-03-09 at 09:27 +0100, Haenggi Stefan wrote:
> Gentlemen,
>
>
>
> I am using an USRP2 with a WBX RF board over the UHD interface to
> receive "COMPLEX_INT16" IQ baseband samples.
>
> It works very well and I am very happy with the performance of the
> device.
>
> Analyzing the received IQ data, I noticed the following :
>
>
>
> 1.) Max IQ receive values
>
> The IQ values (signed 16bit int) seem to clip around -13000 and
> +13000. Is it correct that the IQ values have a maximal
>
> range of about -13000 and +13000 ?
>
Sounds right.
>
>
> 2.) Receive gain of WBX RF board
>
> If a set via UHD the receive RF gain of the WBX on minimum (0dB), the
> IQ values clip @ RF carrier level of -8dBm
Hey, be careful. We don't recommend putting more than -10dBm into the
WBX. That's a pretty dang strong signal for a receiver to take. =)
>
> If a set via UHD the receive RF gain of the WBX on maximum (30dB), the
> IQ values clip @ RF carrier level of -41dBm
>
> Does that seem to be right ? Or can more RF gain be activated
> somehow ?
41-8 = 33. So that's pretty close.
>
> If these levels are right, I will lose a lot of effective IQ bits if I
> have a weak RF input signal, as example -80dBm.
That's right. There's also a bunch of fixed gain in there that you don't
have control over. Problem is, if you add more gain in there, any
additional gain will just magnify the noise; in other words, your SNR is
limited by the noise level, not by quantization.
>
>
>
> 3.) Glitch in Timedomain
>
> I noticed each time I receive a batch of IQ samples (example 1000),
> the value of IQ samples No 15 to 20 seem to
>
> be around 0, regardless what the RF input signal is. This is
> independent of sampling rate.
>
> Is that only my setup ? Or do others see that as well ?
>
Hmm. Typically, when you first start the device, it'll spit out some
weird samples as the RF chain settles down. Can you give more
information? How are you taking samples in?
>
>
> 4.) Overall Receive Phasenoise
>
> Has somebody already measured RMS PhaseJitter of the receive path? In
> which range is that ? Does it change
>
> significantly by locking the USRP2 to a 10MHz reference with low
> phasenoise?
The 100MHz reference oscillator on the USRP2 has crazy-low phase noise,
especially out past a few kHz. Because of this, and because the loop
bandwidth for locking the ref is pretty low, locking to an external
oscillator can improve the close-in phase noise if you have a really,
really quiet sig gen but won't do much for far out noise. Locking to a
10MHz ref primarily provides improvement in frequency accuracy, not
phase noise.
--n
>
>
>
> Thanks a lot for your support
>
> Stefan
>
>
> _______________________________________________
> USRP-users mailing list
> [email protected]
> http://lists.ettus.com/mailman/listinfo/usrp-users_lists.ettus.com
------------------------------
Message: 2
Date: Wed, 9 Mar 2011 10:53:10 -0800
From: yqluo <[email protected]>
To: Mike McLernon <[email protected]>
Cc: "[email protected]" <[email protected]>
Subject: Re: [USRP-users] Data length of USRP2 receiver in Matlab
simulink
Message-ID:
<[email protected]>
Content-Type: text/plain; charset="windows-1252"
Hi, Mike:
I have tried your suggestions on multi-rate FIR filter to change data length
and rate simultaneously. However, when I add the corresponding FIR filter at
the receiver part, the received signal suffers lots of errors.
For example, FIR rate converter with mfilt.firsrc(358,96) is implemented in
sender, and FIR filter with mfilt.firsrc(96,358) is in the receiver (as a
simple case, I do it in simulink system without usrp2 modules). But the
error rate is too high in this structure.
Is there anything wrong with the design of FIR rate converter at the
receiver? Any help will be really appreciated.
On Wed, Feb 2, 2011 at 3:00 PM, Mike McLernon
<[email protected]>wrote:
> Hi Yuanqian,
>
>
>
> Good questions ? see responses below . . .
>
>
>
>
>
> *From:* [email protected] [mailto:[email protected]] *On Behalf Of *yqluo
> *Sent:* Thursday, January 27, 2011 4:38 PM
> *To:* Mike McLernon
> *Cc:* [email protected]
> *Subject:* Re: [USRP-users] Data length of USRP2 receiver in Matlab
> simulink
>
>
>
> Hi, Mike:
>
> Thanks for your quick reply. I have several questions on this technique.
>
> 1. for certain transceiver system, does that mean we need to apply the
> downsampling FIR filter ( 356/578) in the transmitter part, and then use
> upsampling FIR filter (578/356) in the receiver to change the frame length
> back?
>
> I believe the actual numbers are 358 and 576, but you have the idea
> correct.
>
>
>
> 2. do we need to consider about the effect of time-offset of FIR filter?
> since FIR filter in this case should have very large order grade.
>
> It depends on what you want to do with the signal. If you are using it to
> transmit audio, and you don?t care about a small delay, then you don?t need
> to worry about it. If the model needs to transmit digitally modulated
> signals, then you may need to compensate for the effect. However, if you
> have a standalone receiver model, then you will need to perform
> synchronization using only the received data, and the filter delays
> shouldn?t matter.
>
>
>
> 3. how to deal with the situation that the data rates diverse greatly?
> i.e., if we need 1000/356 upsampling, then FIR filter may not work in this
> situation. I think the rebuffering may be the only solution in this case.
>
> If you can avoid the rebuffering, please do. Remember that you can start
> with an arbitrary frame size, and use resampling as necessary to create the
> length 358 frame. I?ve attached a model, mResmplingForUSRP.mdl, that does
> this with the USRP Transmitter block.
>
> Hth,
>
> Mike
>
>
>
>
>
> Really appreciate your help.
>
> Regards,
>
> On Thu, Jan 27, 2011 at 4:41 AM, Mike McLernon <
> [email protected]> wrote:
>
> Hi Yuanqian,
>
>
>
> We empirically found that rebuffering to change frame size in models with
> USRP I/O blocks severely degraded real-time performance. You?ll note in our
> FM demos that we do not perform any rebuffering, so that we can keep the
> model single rate in Simulink.
>
>
>
> However, if you can be flexible on your sample rate, I can show you a
> technique so that you can get a frame size of 576 while still keeping the
> model single rate in Simulink. You would use resampling filters to change
> the size of the frame. Take a look at the FM demos to see how this is done
> ? turn on the display of signal dimensions, and it should become clear. If
> not, I can show you a simple model that converts a frame size of 358 to one
> of 576 while keeping the Simulink model single rate.
>
>
>
> Best,
>
> Mike
>
>
>
>
>
> *From:* [email protected] [mailto:
> [email protected]] *On Behalf Of *yqluo
> *Sent:* Wednesday, January 26, 2011 8:27 PM
> *To:* [email protected]
> *Subject:* [USRP-users] Data length of USRP2 receiver in Matlab simulink
>
>
>
> Hi, everyone:
>
> I have tried to modify the transceiver parts of USRP2 model in the Matlab
> simulink environment. I want to change the receiver data length to 576
> instead of 358 to match the receiver design. But after changing the value,
> there is no signal received at all.
>
> Here is what i did:
> 1. Disable link at "Link option", and then break the link. (in order to
> change the length value)
> 2. change the packet length and data length to 576.
>
> I am not sure whether the received data length in the USRP2 receiver model
> can be changed or not? Is there any user manual to show how to modify the
> USRP2 model in simulink?
>
> Regards,
>
>
>
>
> --
> Yuanqian Luo
> Department of Electrical and Computer Engineering
> University of Victoria
>
>
>
>
> --
> Yuanqian Luo
> Department of Electrical and Computer Engineering
> University of Victoria
>
--
Yuanqian Luo
Department of Electrical and Computer Engineering
University of Victoria
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
<http://lists.ettus.com/pipermail/usrp-users_lists.ettus.com/attachments/20110309/40857b72/attachment-0001.html>
------------------------------
Message: 3
Date: Wed, 9 Mar 2011 11:05:34 -0800
From: Jason Abele <[email protected]>
To: Alexandru Csete <[email protected]>
Cc: [email protected]
Subject: Re: [USRP-users] UHD mirror stuck
Message-ID: <20110309190534.GQ2526@manzanita>
Content-Type: text/plain; charset=us-ascii
> The UHD mirror on Github seems to be stuck at Feb 16/17.
Fixed, thanks
------------------------------
Message: 4
Date: Wed, 9 Mar 2011 12:10:04 -0800
From: zhe yang <[email protected]>
To: [email protected]
Subject: Re: [USRP-users] Welcome to the "USRP-users" mailing list
Message-ID:
<[email protected]>
Content-Type: text/plain; charset="iso-8859-1"
Hi,
I am new to USRP2 and I am playing with simulink+UHD now. I am wondering is
there any demo about simulink+UHD+USPR2, such as
a simplest transmitter and receiver. Thanks.
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
<http://lists.ettus.com/pipermail/usrp-users_lists.ettus.com/attachments/20110309/5a8c5f58/attachment-0001.html>
------------------------------
Message: 5
Date: Wed, 9 Mar 2011 21:19:53 +0000
From: Mike McLernon <[email protected]>
To: zhe yang <[email protected]>
Cc: "[email protected]" <[email protected]>
Subject: Re: [USRP-users] Welcome to the "USRP-users" mailing list
Message-ID:
<e3879be9a282cb45aab7ce258a9ae48f071...@exmb-01-ah.ad.mathworks.com>
Content-Type: text/plain; charset="us-ascii"
A UHD-based Simulink block that interfaces with the USRP2 is not yet shipping,
but we are working to release that capability soon. Simulink does have
UDP-based blocks that ship in R2010a and R2010b. So if you want to use
Simulink, you'll need UDP. If you want to use UHD, you'll need to use GNU
Radio for now.
Mike
From: [email protected]
[mailto:[email protected]] On Behalf Of zhe yang
Sent: Wednesday, March 09, 2011 3:10 PM
To: [email protected]
Subject: Re: [USRP-users] Welcome to the "USRP-users" mailing list
Hi,
I am new to USRP2 and I am playing with simulink+UHD now. I am wondering is
there any demo about simulink+UHD+USPR2, such as
a simplest transmitter and receiver. Thanks.
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
<http://lists.ettus.com/pipermail/usrp-users_lists.ettus.com/attachments/20110309/8194582e/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 17
*****************************************
