Re: [Discuss-gnuradio] IEEE 802.11 Transmitter in X310s

[Bastian Bloessl]

Hi,

On 04/09/2016 12:29 AM, Abhinav Jadon wrote:
> Hi,
> My setup consists of two X310s - one transmitter and one receiver -
> connected with SMA cable.
> The receiver code has been independently tested by capturing the AP packets.
> The transmitter code was tested in a software loopback with the receiver.
> The gains on both sides were adjusted such that  there was no clipping.
> attenuator (-10db) were used to connect the two USRPs over SMA link.
>
> The transmitter transmits the packet but at the receiver end, these
> packets are not successfully decoded. A deeper look reveals that for a
> high correlation value (0.8) , packets are being detected by the packet
> detection mechanism but the signal field is decoded wrongly in all the
> packets detected. . Since unit testing of every block was successful, it
> was my hypothesis that the source of error lies on the
> transmitter-hardware side.
>
> I began scraping thorough the mail archives and got hold of this :
> http://lists.ettus.com/pipermail/usrp-users_lists.ettus.com/2015-September/015576.html
>
> This thread details on the problem related to burst transmissions in
> X310. Apparently, the successive bursts are not same - the end of first
> burst is appended to the second burst.
>
> I don't have the hardware to verify the claim, but assuming that it is
> correct. It still doesn't explain the wrong decoding of the signal field.


To avoid this and to give the device some time to stabilize at the 
beginning of a burst I usually pad the frame with zeros (Packet Pad block).

Also you could try to change the LO offset option (at sender and 
receiver) to tune the oscillator out of your band of interest.

Best,
Bastian