USRP Software Radio

cancel
Showing results for 
Search instead for 
Did you mean: 

Trying to record a WiFi signal with niUSRP EX One Shot RX.vi and with a NI-USRP 2932 Unit

Hello,

 

I am currently trying to capture and record an 802.11 WLAN signal from an actual WLAN Router I have in my testbed. My NI-USRP model is an 2932 and I have it say 5 feet away from the router.

 

It's a NETGEAR WLAN Router and I am certain there are no other WiFi signals around since I have scanned all other frequency channels. The only signal that is being transmitted is the one from the Router I have which I have set to Channel 1:  2.412GHz.

 

I do see the signal from the router but it seems to have some sort of carrier frequency mismatch or I really don't know what it is.

 

On the other hand, If I transmit something with another NI-USRP (I also have an 2920 unit, and of course I transmit at a lower frequency range say 2.0GHz so that  both USRP units can operate in that frequency range), any type of waveform say a tone or anything, because of the proximity of the units, i do get the signal and there is no such mismatch or error. Of course there is some noise, etc, but not like this.

 

So when I try to receive the signal from the Router to my NI-USRP 2932 unit, I get this type of raw signal when using the niUSRP EX One Shot RX.vi example:

 

ScreenHunter_12 Sep. 19 16.14.jpg

 

You can see that it's clearly where the WLAN Frame starts because of the difference in power. It is very clean signal as well, but I see those sinusoidal-like lobes all over the signal.

Any ideas?

 

Thanks

Regards,

Erick

0 Kudos
Message 1 of 5
(5,582 Views)
Those are caused by frequency offset. A very minimal frequency offset causes low freq sinusoids. The complex magnitude will look very nice like constant envelope.
0 Kudos
Message 2 of 5
(5,579 Views)

ErikL,

 

Thanks for the quick reply. A few more questions:

 

-Frequency Offset even if the Tx Router and the Rx USRP2932 are in line of sight and like 3-4 feet away?

 

-How come I don't see a frequency offset when I transmit from another Tx USRP2920 some OFDM Symbols? i.e. How come I don't see the same sinusoid shapes on the Rx USRP2932 when I transmit from an Tx USRP2920 at same distance and LOS than when I do with case Tx Router and Rx USRP2932? (When using case TxUSRP2920/RxUSRP2932, both USRP units Tx and Rx are not synchronized, they are in fact in different computers and different LabVIEWs. I am not trying to synchronize, rather just look at the signal at the Rx USRP2932 in some shape similar to the one on the Tx USRP2920)

 

 

Regards

Erick

0 Kudos
Message 3 of 5
(5,574 Views)
If you transmit a single sinusoid you are in fact transmitting a single tone at a frequency offset, so it looks the same, except it is only one at a time. Likely you are receiving the BPSK beacons from your router and not OFDM packets. In either case you would have to adjust for frequency offset. Since most radios use inexpensive TCXO crystals there is almost always frequency accuracy due to clock drift or Doppler offsets to deal with. Cell phones discipline their clock to the carrier from the tower. Gps discipline themselves to the satellites. In wifi it's done as carrier recovery. ... There are many aspects like this that don't get full coverage in the textbooks. Most research papers start with "assume the system is synchronized".... Which is a pretty big assumption given time, frequency, and even sampling differences between tx and rx. Wifi OFDM gives you a preamble on each packet to address course freq offset, sync, and multipath and the cyclic prefix on the OFDM symbol for fine corrections.
0 Kudos
Message 4 of 5
(5,563 Views)

Thank you ErikL,

 

I do understand the steps such asi signal acquisition, symbol timing, coarse and fine frequency offset correction, channel estimation, channel compensation, phase tracking and other steps involved in an 802.11 OFDM receiver. But this signal doesn't seem to look anywhere close to a short or long training sequence to actually apply these corrections to it. It seems to be a repeating pattern in fact.

 

Here are the RAW signals I am receiving (I and Q plotted on graph):

 

From Router (@2.412GHz):

ScreenHunter_12 Sep. 19 16.14.jpg

 

ScreenHunter_16 Sep. 20 15.55.jpg

 

 

From another USRP (@2GHz):

ScreenHunter_17 Sep. 20 16.31.jpg

 

 ScreenHunter_18 Sep. 20 16.32.jpg

 

 

As you can see, from the USRP, the signal is discernable and we can actually see the initial preamble. It doesn't look like a sinusoid but still it does have a frequency offset.

 

I have in fact created an OFDM signal in MATLAB and artificially introduced a carrier offset to it and it doesn't look like the received signal from the router.

 

Any ideas of what might be happening? Is the NI USRP2932 capable of receiving 802.11 OFDM signals by means of SDR or is there another hardware step that can't be done in SDR? Or what is the difference between transmission from an 802.11 Router and an NI-USRP unit?

 

Regards

Erick

0 Kudos
Message 5 of 5
(5,555 Views)