Hi ,
I am trying to measure the RSSI on an usrpn200 which acts as a receiver.
i
have used the rx_ascii_dft.py and uhd_fft.py both and i have made a
host of
measurements but the received power is not reliable for different
transmit
powers at different amplitude of 0.25,0.50 and 0.85 (which i took
randomly ) … the channel attenuation is not less for more transmit
power in
short …
Is there any other way to measure the RSSI values besides these two apps
uhd_fft.py and rx_ascii_dft.py ?i have run a set of measurements and the
resultant received powers are not reliable at all ,according to me . i
am
still an amateur regarding this .
how about running the rssi.v verilog file ? … will i be able to extract
the
rssi values directly from the fpga ?
Anticipating your feedback in earnest
Bonee
On 22/10/11 10:56 PM, Bonee S. wrote:
Hi Marcus ,
Sorry i did not explain to you my set-up , i have two usrpn200’s with
RX2400 as the daughterboards . i have omnidirectional antennas . Its a
line of sight experiment . i used .benchmark_tx.py -f 2475M -m bpsk
–from-file=test.pdf -S 2 -s 50 --tx-amplitude=0.25 -v
and on the receiver side i ran ./uhd_fft.py and ./rx_ascii_art_dft.py .
the problem i get in uhd_fft.py is that , as i vary the transmit
powers from 0.25 to 0.85 , i am getting -20 dB constant peak at a
fixed distance . logically it should increase as you said but i am
getting constant received values .
My suspicion, having not looked at the code, is that the magnitude of
the signals arriving at the
“amp” block in that flow-graph are already sufficiently larger than
“1”, that you’re seeing clipping
somewhere in the chain.
You could read the RX-side aux_adc, but it would give you mis-leading
results, since the power-detector
is showing you the power across the entire post-mixer bandwidth. It
really, really is better to calculate
the RSSI from the incoming data, and since you’d have to modify your
application anyway to
read the aux_adc, you might as well add a couple of blocks to
calculate received power properly.
Also, not all cards have an RSSI detector available in hardware, so
assuming that they do is
non-portable to other types of cards.
On 22/10/11 09:41 PM, Bonee S. wrote:
and the resultant received powers are not reliable at all ,according
to me . i am still an amateur regarding this .
how about running the rssi.v verilog file ? … will i be able to
extract the rssi values directly from the fpga ?
Anticipating your feedback in earnest
Bonee
Received power is proportional to the average of the square of the I and
Q voltages. If you
use a Complex-to-Mag**2 block, followed by a low-pass filter, you’ll
get an (uncalibrated!!)
received-power estimate.
An increase in voltage magnitude from 0.25 to 0.5 should create a
received power increase of
6dB, since doubling the voltage increases the power by a factor of 4
(6dB), assuming a reasonably
linear channel between TX and RX.
You haven’t said what your test setup is: antennae, daughtercards, etc.
What are you using to transmit? Is it another USRP device, or something
else?
–
Principal Investigator
Shirleys Bay Radio Astronomy Consortium
Hi,
I was reading your very interesting discussion and I was wondering if
you can give us some tips on how to compute the RSSI by software and
hence the blocks that we should add.
Tnx
Vanessa
On Sun, Oct 23, 2011 at 3:35 PM, Marcus D. Leech [email protected]
wrote:
logically it should increase as you said but i am getting constant received
Please tell me what to do next . i read on your site that , you can measure
results, since the power-detector
–
Vanessa GARDELLIN, Ph.D.
Researcher
Institute for Informatics and Telematics (IIT),
Italian National Research Council (CNR)
Via G. Moruzzi 1
56124 Pisa - ITALY
Phone: +390503153267
Room: B65/c
E-mail: [email protected]
WWW: http://www.iit.cnr.it/staff/vanessa.gardellin/
Skype: gardellin.vanessa
How can I read the output of keep-one-in-N?
Thank you
V.
On Wed, Oct 26, 2011 at 5:32 PM, Marcus D. Leech [email protected]
wrote:
RSSI estimates should vary slowly, so you want to set the “alpha” value on
the appropriately, and then the sample rate can
be quite low, relative to the sample rate of the main sample stream.
–
Vanessa GARDELLIN, Ph.D.
Researcher
Institute for Informatics and Telematics (IIT),
Italian National Research Council (CNR)
Via G. Moruzzi 1
56124 Pisa - ITALY
Phone: +390503153267
Room: B65/c
E-mail: [email protected]
WWW: http://www.iit.cnr.it/staff/vanessa.gardellin/
Skype: gardellin.vanessa
On 26/10/2011 12:15 PM, Vanessa Gardellin wrote:
How can I read the output of keep-one-in-N?
Thank you
V.
Like any other signal flow. If you need it to appear in a variable, you
can use a “probe”.
On 26/10/2011 11:27 AM, Vanessa Gardellin wrote:
Hi,
I was reading your very interesting discussion and I was wondering if
you can give us some tips on how to compute the RSSI by software and
hence the blocks that we should add.
Tnx
complex_to_mag**2–>single-pole-iir-filter—>keep-one-in-N
RSSI estimates should vary slowly, so you want to set the “alpha” value
on the appropriately, and then the sample rate can
be quite low, relative to the sample rate of the main sample stream.