Re: [Discuss-gnuradio] BBN 802.11 bbn_80211b_rx.py

[Mohammad Hamed Firooz]

802.11b standard uses 11 chip barker code for spreading. It means each 
bit is converted to 11 bits and for DBPSK and QBPSK modulation it leads 
to a signal with 11MHz bandwidth. When you downsample data with 8 (-d 
8) it means you cut the frequency to 4Mhz (instead of 11MHz) and it 
means you reduce your signal strength.

BBN transmitter code doesn't use barker spreading. They use raised 
cosine and its spreading factor (or gain) is less than 11. So receiver 
doesn't have to cut the frequency and it works fine.

Did I get your question right?

Quoting yyzhuang <address@hidden>:

> But when I use one usrp for tx and another for rx, no need to specify "-d 8".
> Could you tell me why is this? Thanks
>
>
> Mohammad Hamed Firooz wrote:
> > 802.11b standard use barker spreading, So if you want to receive data
> > from a standard 802.11b transmitter you have to use barker. For
> > decimation it should be as high as possible because 802.11b receiver
> > needs more more than 22M complex sample per sec to work correctly. But
> > USB cannot handle that much bit-rate so we have to downsample it to
> > 8McSps. This reduces SNR which reduces receiver performance and range
> > of activity.
> >
> > regards,
> > hamed
> >
> > Quoting yyzhuang <address@hidden>:
> >
> > > Thanks.
> > >
> > > When I use "./bbn_80211b_rx.py -f 2.44G -v -b ", no packets get printed
> > > out.
> > > But the output of "./bbn_80211b_rx.py -f 2.462G -d 8 -b -v " is normal,
> > > we
> > > got a bunch of packets:
> > >
> > > Bits Per Encoded Sample = 8
> > > adc frequency =  64000000
> > > decimation frequency =  8
> > > input_rate =  8000000
> > > gain =  45.0
> > > desired freq =  2462000000.0
> > > baseband frequency 2460000000.0
> > > dxc frequency -2000000.0
> > > Samples per data bit =  8
> > > > > > gr_fir_ccf: using SSE
> > > PKT: len=77, rssi=-22, src=00:13:46:16:28:AA, time=557560, rate=1 Mbps
> > > PKT: len=77, rssi=-29, src=00:13:46:16:28:AA, time=1786312, rate=1 Mbps
> > > PKT: len=77, rssi=-26, src=00:13:46:16:28:AA, time=1991192, rate=1 Mbps
> > > ...........
> > >
> > > I checked the code:
> > >
> > > parser.add_option("-d", "--decim", type="int", default=16,
> > >                          help="set fgpa decimation rate to DECIM
> > > [default=%default]")
> > > parser.add_option("-b", "--barker", action="store_true",
> > >                          default=False,
> > >                          help="Use Barker Spreading [default=%default]")
> > >
> > > -d 8 is setting the fpga decimation rate to 8, and -b means Use Barker
> > > Spreading. How should we know what decimation rate and spreading to use?
> > > Thanks.
> > >
> > > BTW, if one ursp for tx and one for rx, we dont have to specify these
> > > options:
> > > on the sending side: ./bbn_80211b_rx.py -f 2.44G -v -b
> > > Using TX d'board A: Flex 2400 Tx MIMO B
> > > > > > gr_fir_ccf: using SSE
> > > spb:       8
> > > interp:   32
> > > (spb = samples/baud, default 8; interp = fpga interpolation rate, default
> > > 32
> > > )
> > >
> > > on the receiving side: ./bbn_80211b_tx.py -f 2.44G
> > > Bits Per Encoded Sample = 8
> > > adc frequency =  64000000
> > > decimation frequency =  16
> > > input_rate =  4000000
> > > gain =  45.0
> > > desired freq =  2440000000.0
> > > baseband frequency 2436000000.0
> > > dxc frequency -4000000.0
> > > Samples per data bit =  8
> > > > > > gr_fir_ccf: using SSE
> > > PKT: len=1477, rssi=-10, src=20:61:6e:64:20:73, time=16024824, rate=1
> > > Mbps
> > > PKT: len=1477, rssi=-12, src=20:61:6e:64:20:73, time=16036920, rate=1
> > > Mbps
> > > PKT: len=1477, rssi=-11, src=20:61:6e:64:20:73, time=16049016, rate=1
> > > Mbps
> > > PKT: len=1477, rssi=-13, src=20:61:6e:64:20:73, time=16061112, rate=1
> > > Mbps
> > >
> > > Thanks!
> > >
> > >
> > > Brian Padalino wrote:
> > > > On Mon, Jul 28, 2008 at 12:28 PM, yyzhuang <address@hidden> wrote:
> > > > > What does "combination of the LO in the front end and the digital
> > > > > downconverter" mean please?
> > > >
> > > > For general radio architecture, please see:
> > > >
> > > >     http://en.wikipedia.org/wiki/Superheterodyne
> > > >
> > > > Generally, the LO/synthesizer has some discrete step size (12.5kHz,
> > > > 250kHz, etc).  To get down to baseband, the resultant signal is then
> > > > shifted in frequency down using the digital downconverter:
> > > >
> > > >     http://en.wikipedia.org/wiki/Digital_Down_Converter
> > > >
> > > > Brian
> > > >
> > > >
> > > > _______________________________________________
> > > > Discuss-gnuradio mailing list
> > > > address@hidden
> > > > http://lists.gnu.org/mailman/listinfo/discuss-gnuradio
> > >
> > > --
> > > View this message in context:
> > > http://www.nabble.com/BBN-802.11-bbn_80211b_rx.py-tp18642793p18700741.html
> > > Sent from the GnuRadio mailing list archive at Nabble.com.
> > >
> > >
> > >
> > > _______________________________________________
> > > Discuss-gnuradio mailing list
> > > address@hidden
> > > http://lists.gnu.org/mailman/listinfo/discuss-gnuradio
> >
> >
> >
> > ----- End forwarded message -----
> >
> >
> > --
> > Mohammad Hamed Firooz
> > ECE Department, U of Utah
> > Salt Lake City
> > UT, 48112
> >
> >
> > _______________________________________________
> > Discuss-gnuradio mailing list
> > address@hidden
> > http://lists.gnu.org/mailman/listinfo/discuss-gnuradio
>
> --
> View this message in context: 
> http://www.nabble.com/BBN-802.11-bbn_80211b_rx.py-tp18642793p18701054.html
> Sent from the GnuRadio mailing list archive at Nabble.com.
>
>
>
> _______________________________________________
> Discuss-gnuradio mailing list
> address@hidden
> http://lists.gnu.org/mailman/listinfo/discuss-gnuradio



--
Mohammad Hamed Firooz
ECE Department, U of Utah
Salt Lake City
UT, 48112