Re: [Discuss-gnuradio] Please Test N210 Users

[Marcus Müller]

Ok, SangHyuk,

we've told you *so many times* now that there are *millions* of ways for a communication link to fail, that it's impossible to tell remotely. Also, I mean, really, what about --rx-gain? Why don't you specify that?

Also, we repeatedly pointed out that benchmark_rx/_tx are both outdated, and not an appropriate way of understanding what goes "wrong" on your channel.

Please stop using these. You should really /analyze/ what goes wrong on your channel, using common digital communication knowledge that you probably have acquired over the last 5 months that you've been working on this project.

As usual, and as already explained, what about sending out a reference signal, and comparing what you receive in spectrum domain with what you've sent. What about interference? are your antennas, gains, bandwidths suitable?

Best regards,

Marcus

On 08.07.2016 07:50, SangHyuk Kim wrote:

Hi, 

I have used USRP N210 with CBX40 at 1.18GHz

I got a low performance when two USRP N210 communicate

I used 'benchmark_tx.py' and 'benchmark_rx.py' located at ~/gnuradio/gr-digital/example/ofdm

Commands I used are below each other :

Tx) ./benchmark_tx -f 1.18G -m qpsk --tx-gain=31.5 -W 2500000

Rx) ./benchmark_rx -f 1.18G -m qpsk -W 2500000

(other options are default : packet size = 400, fft = 512, occupied toens = 200, etc.)

And I counted how many packets are received at Rx machine

Then I calculated throughput on Rx during 60 sec with below equation :

Throughput[bps] = (# of total packet - # of error packet) * 8 * 400 / 60

I derived about 1 Mbps at Rx and I think it is too poor

Am I something missing ?

So, I ask USRP N210 users to try that and tell me how about yours.

Thanks.

_______________________________________________
Discuss-gnuradio mailing list
address@hidden
https://lists.gnu.org/mailman/listinfo/discuss-gnuradio