Matt, Thanks for clarifying the matter. My colleague discussed it with you and we look forward to receive the correct antennas. Regards -Faisal Just to clarify for everyone, since I get a LOT of calls about this...
on 2007-04-07 21:13
on 2009-06-16 16:33
Hi We bought 2 USRP boards very recently and it doesn't seem to work. We have daughter boards XCVR 2450 and vertical antennas (Black colored) VERT2450. We also have TVRX daughter board which works properly. We had been trying usrp_siggen.py, usrp_oscope.py on each of these. Also digital/benchmark_tx.py and benchmark_rx.py on both of these. 1) usrp_oscope.py doesn't give me a proper sine wave at all! 2) usrp_siggen.py or even usrp_fft.py doesnt seem to give me satisfactory results. (I am comparing the results according to the screen shots i saw in the web page http://staff.washington.edu/jon/gr-osx/gr-osx-usrp...) 3) benchmark_tx/rx do not receive any packets though i give a transmit amplitude of 30000 and rx-gain of 75. The transmission seems to be sending 666 packets where as i receive only total of around 60 packets on the receiver end with ok=FALSE Always!! I get a 100% error rate. Do you think there are any problems with the daughterboards? or the combination of daughter board with the antennas? Kindly let me know how do i check the functioning of the daughter boards and/or the antennas! Kindly help Udaya Syed Faisal Shah wrote: > >> it labeled "Highgain Antennas 2.4-2.48 GHz". See the datasheet for the >> which came with it. > > -- View this message in context: http://www.nabble.com/High-packet-error-and-recept... Sent from the GnuRadio mailing list archive at Nabble.com.
on 2009-06-16 17:27
On Tue, Jun 16, 2009 at 7:31 AM, udadidd<firstname.lastname@example.org> wrote: > 3) benchmark_tx/rx do not receive any packets though i give a transmit > amplitude of 30000 and rx-gain of 75. The transmission seems to be sending > 666 packets where as i receive only total of around 60 packets on the > receiver end with ok=FALSE Always!! I get a 100% error rate. Not commenting on the rest, but with respect to power and gain settings: More transmit power is not always better. More receiver gain is not always better. The above settings are almost guaranteed to fail. You should use usrp_fft.py on the receiver side and leave it at mid-gain. Then use benchmark_tx.py and increase or decrease the transmit amplitude, while monitoring the receive spectrum, until you get the widest separation between the noise floor and the modulated signal spectrum. In general, the RFX2400 and other RFX boards start to go into gain compression at a transmit amplitude of about 10-15% DAC full-scale, which with the USRP1, is about 3000-5000. Beyond this, transmitting non-constant envelope signals will result in significant distortion. For a bench top transmission channel, the RFX2400 can generate far more power than is needed to get essentially error-free reception. Once you have a transmit power that corresponds to the "best looking" spectrum in usrp_fft.py on the receiver, you can then adjust the receiver gain to further improve the signal-to-noise ratio. Too low a gain doesn't amplify the signal enough, so you see it drop into the noise floor. Too high a gain starts to increase the noise floor without further increasing the signal level, due to non-linearities in the receiver causing distortion. So again, there will be a "just right" spot. At this point, you may be able to go back to the transmitter and further improve things now that the receiver is operating with the better gain setting. Once you have a good pair of transmit amplitude and receiver gain for your particular setup, then you can start running the benchmark_rx.py script with the same gain setting to start receiving the packets being sent by the transmitter. Johnathan