Re: [Discuss-gnuradio] problem with 2Mbps of BBN code(802.11b)

[Michael Gray]

Since you are going from B-PSK to Q-PSK modulation you may have a problem 
with spectral inversion.  With B-PSK it won't matter, however with Q-PSK a 
+90 degree phase change will look like -90 degrees at the receiver.

You can invert the spectrum with the set_rx_freq command.  Just set a 
negative IF frequency.

src.set_rx_freq (0, -IF_freq)

Hope that helps....



On Sun, 11 Nov 2007, Mohammad Hamed Firooz wrote:

> 
>     Hi
> I hope at least one of the members of BBN group read this email.
> I used BBN implementation of IEEE802.11b to receive packets from
> commercial products. So I always use Barker code as the spreading key.
> As it had been reported before, Their code works with 1Mbps and
> doesn't work with 2Mbps. I cannot understand why. Since bandwidth of
> DBPSK and DQPSK are the same, the problem is not because of low
> sampling rate. I went precisely through thier code step by step and
> everything seems to be correct. In fact, there is a little difference
> between 1Mbps and 2Mbps modes.
> To explain the problem clearly I copy and paste results of receiving
> preamble and header of the short PLCP. You may know that for short
> PLCP, preamble is sent by 1Mbps while Header is transmitted using
> DQPSK (2Mbps) and both of them uses Barker code.
> 
> Recieved (short) header!
>     signal: 0xBA
>     service: 0xE7
>     length: 0x640D
>     crc: 0x388A
> Calculated crc: 0x5112
> Recieved (short) header!
>     signal: 0xFF
>     service: 0xCB
>     length: 0x1A50
>     crc: 0x2C29
> Calculated crc: 0x761C
> Recieved (short) header!
>     signal: 0xF9
>     service: 0x11
>     length: 0x8340
>     crc: 0x8B78
> Calculated crc: 0x5D2A
> Recieved (short) header!
>     signal: 0x31
>     service: 0x1F
>     length: 0xC729
>     crc: 0xA902
> Calculated crc: 0x255D
> Recieved (short) header!
>     signal: 0x04
>     service: 0x09
>     length: 0x9170
>     crc: 0xBD57
> Calculated crc: 0x67E8
> 
> Almost never header is received correctly. I am looking for the reason
> of the bug. As I mentioned before it is not because of low sampling
> rate (the sampling rate is 8MSPs)
> 
> regards,
> hamed
> 
> 
> 
> 
> _______________________________________________
> Discuss-gnuradio mailing list
> address@hidden
> http://lists.gnu.org/mailman/listinfo/discuss-gnuradio
>