Re: USRP 2.0

Matt speaks of it in some detail here:
http://dewy.fem.tu-ilmenau.de/CCC/CCCamp07/video/m4v/cccamp07-en-2012-GNU_Radio_the_Universal_Software_Radio_Peripheral.m4v

No mention of the final price, unfortunately.

I am also eagerly awaiting USRP2. So any new information about it
would be welcome.

Has anyone experimented with the digital compensation of the I/Q
imbalance from direct conversion asics like the one in the dbsrx? I’m
interested in the algorithms employed and how successful the
compensation can be.

Thanks,
Clark

This question has nothing related with the subject title “USRP 2”

regards,

Firas

Clark P. wrote:

Your smile counts. The more smiles you share, the more we donate. Join
in.
www.windowslive.com/smile?ocid=TXT_TAGLM_Wave2_oprsmilewlhmtagline


Discuss-gnuradio mailing list
[email protected]
http://lists.gnu.org/mailman/listinfo/discuss-gnuradio


View this message in context:
http://www.nabble.com/USRP-2.0-tp14127248p14235737.html
Sent from the GnuRadio mailing list archive at Nabble.com.

Has anyone experimented with the digital compensation of the I/Q
imbalance from direct conversion asics like the one in the dbsrx? I’m
interested in the algorithms employed and how successful the
compensation can be.

It depends on how detailed a model you’re willing to
implement and the stability of the hardware (temperature
is usually the biggest issue). I haven’t looked at the
dbsrx, but I’ve done something similar for a couple
of transmitter designs.

In rough numbers, a quadrature modulator or demodulator
might be natively good to around 40 dB of image rejection.
Fixing the LO and applying fixed gain and phase corrections
might improve this to 50 dB. The deluxe model would go
something like: take your complex-baseband signal; take
its conjugate; apply an n-tap complex filter for some n,
which includes a scale factor (of around -40 dB) and
a frequency-dependent amplitude and phase response; and
finally add it back to your original signal. If the
filter is right (it would be obtained from some sort of
calibration), it will serve to cancel the image response.
Image rejection of 65–70 dB is attainable over
nontrivial bandwidths.

If you can somehow close the loop with a pilot signal,
even better rejection might be had with an adaptive
canceller continuously estimating the filter mentioned
above, tracking out the last smidgen of temperature-dependent
bumps and wiggles. For a receiver, this could be something
like injecting a comb at a suitable level prior to the
receiver, then watching the image frequencies for any
tone power, and twiddling the filter to cancel it.
Oh, and subtract out the comb before it hits the real
receiver. (Pseudonoise could also serve, but it might
be less convenient to generate.)

Cheers,
Peter

Date: Sun, 9 Dec 2007 04:58:36 -0800> From: [email protected]> To: [email protected]> CC: [email protected]> Subject: Re: [Discuss-gnuradio] direct conversion compensation> > > >> > Has anyone experimented with the digital compensation of the I/Q > > imbalance from direct conversion asics like the one in the dbsrx? I’m > > interested in the algorithms employed and how successful the > > compensation can be.> > It depends on how detailed a model you’re willing to> implement and the stability of the hardware (temperature> is usually the biggest issue). I haven’t looked at the> dbsrx, but I’ve done something similar for a couple> of transmitter designs.> > In rough numbers, a quadrature modulator or demodulator> might be natively good to around 40 dB of image rejection.> Fixing the LO and applying fixed gain and phase corrections> might improve this to 50 dB. The deluxe model would go> something like: take your complex-baseband signal; take> its conjugate; apply an n-tap complex filter for some n,> which includes a scale factor (of around -40 dB) and> a frequency-dependent amplitude and phase response; and> finally add it back to your original signal. If the> filter is right (it would be obtained from some sort of> calibration), it will serve to cancel the image response.> Image rejection of 65–70 dB is attainable over> nontrivial bandwidths.> > If you can somehow close the loop with a pilot signal,> even better rejection might be had with an adaptive> canceller continuously estimating the filter mentioned> above, tracking out the last smidgen of temperature-dependent> bumps and wiggles. For a receiver, this could be something> like injecting a comb at a suitable level prior to the> receiver, then watching the image frequencies for any> tone power, and twiddling the filter to cancel it.> Oh, and subtract out the comb before it hits the real> receiver. (Pseudonoise could also serve, but it might> be less convenient to generate.)> > Cheers,> Peter>

Sounds reasonable. It’s still not clear to me if the imbalance is just
at DC (because that is where the LO is) or if it varies across the band.
For example, if I have a 6 MHz bandwidth that I direct convert to
baseband can I tune a narrow band DDC (say 30 kHz) around that bandwidth
without the performance degradation?

Your answer implies the imbalance is across the band but I’ve seen other
references that say to ac couple the ADCs or in the case of OFDM the
center bins are nulled which makes me think the rest of the band doesn’t
suffer as much from the imbalances?

Thanks,
Clark

[ I/Q compensation ]

There’s some discussion in Razavi’s “RF Microelectronics”, section
5.2.2,
and Kenington’s “Software Defined Radio”, section 5.3.1.1. The
distinction
you mention between DC and AC might come from this LO self-mixing
business, a nonlinear effect, and indeed only at DC since it involves
the LO mixing with itself at some random amplitude and phase. AC
coupling would eliminate this artifact, but it would do nothing for the
linear I/Q mismatch errors.

Maybe future versions of the dbsrx could have I/Q compensation
parameters
in the EEPROM. Or is something there already?

Cheers,
Peter

For example, if I have a 6 MHz bandwidth that I direct convert to
baseband can I tune a narrow band DDC (say 30 kHz) around that
bandwidth without the performance degradation?

If you tune the DDC to, say, +500 kHz, then the output of the DDC will
be
contaminated with a small amount of signal from -500 kHz, even though
the DDC filter itself may have very large rejection at -500 kHz. The
damage
is already done pre-DDC, so it won’t vary with the DDC output bandwidth.

With a narrow bandwidth, though, a correction consisting of just scalar
differential-amplitude and differential-phase will be very accurate.
These
values will vary slowly across the 6 MHz.

Your answer implies the imbalance is across the band but I’ve seen
other references that say to ac couple the ADCs or in the case of OFDM
the center bins are nulled which makes me think the rest of the band
doesn’t suffer as much from the imbalances?

Hmm, no, it would seem imbalances still matter at AC. For example,
if the ADC for the I arm had gain 1 dB higher than the ADC for the Q
arm,
one would expect a large imbalance at all frequencies. AC coupling
might
paper over the DC imbalance, but at 1 MHz, say, the I and Q still are
imbalanced
by 1 dB.

Cheers,
Peter

On Sun, Dec 09, 2007 at 07:47:37PM -0800, Peter M. wrote:

Maybe future versions of the dbsrx could have I/Q compensation parameters
in the EEPROM. Or is something there already?

Cheers,
Peter

It’s on the list of things to do. Nobody’s gotten to it yet.
It sounds like you may be the person we’ve been looking for :wink:

There is room in the daughterboard EEPROM for storing the parameters.
We can work out a specific location when you’re ready.

If you’ve got the time and inclination, please have a go at the I/Q
compensation. It would be great to have. Please post any questions,
etc.

Eric

This forum is not affiliated to the Ruby language, Ruby on Rails framework, nor any Ruby applications discussed here.

| Privacy Policy | Terms of Service | Remote Ruby Jobs