RFSA - IQ acquisition

[Carlos A]

Hello,

I'm having problems when acquiring IQ signals with the PXI-5661. In my first test, I used a 1GHz CW tone generated with a PXI-5651, and I was supposed to obtain a constant level on the IQ received signal (i.e. I=1 Q=0), but I get a sine wave on both I and Q components of around 4 KHz with a fase offset of 90°.

When I got this problem, I replaced the 5651 with a HP CW signal generator, but obtained the same result (this time with a 1 KHz sine on I and Q components).

Then, I performed a test with the OOK and PSK modulations of the 5651, but the results didn't correspond with the spected received signal for I and Q:

*the baseband discrete signal spected was multiplied with a 4 KHz tone (see figure for PSK).

*the Q component, wich is expected to be 0 in a binary modulation (OOK,PSK) is the same I component with a tone offset of 90°.

This is a very big problem, because I can't demodulate and recover the PRBS I'm transmiting.

I used the RFSA Getting Started IQ.vi example for the capture. 

What am I doing wrong with my system?

Is the 4KHz tone a product of the multiple downconvertions in the RFSA?  Why it appears when I generate a tone in the transmitter (when I and Q components should be constant)?

Please help me understand this problem.

Thanks in advance.

Carlos A.

[loula]

Hi Carlos,

The PXI-5651 is a CW Source and it can be used for FM, 2-FSK and OOK modulation.

We are trying to reproduce your issue. What's the example you are using to generate with 5651?

Best Regards,

Alexsander Loula

[Carlos A]

Hello, and thanks for your answer.

I was using the "RFSG 565x Digital Modulation.vi" example, but i also made two VIs for the PXI5651 and the PXI5661 trying to solve the problem, but I have the same problem...

I send you the VIs, and some pictures of the results I'm obtaining for a tone with no modulation and with no signal at the analizer input.

Thank you again,

Carlos Viteri

[Carlos A]

...

The pictures...

I forgot to tell you that I have the same problem with OOK and FSK.

Regards,

Carlos Viteri

[wmarinho]

Hello Carlos,

What is your LabVIEW version, modules and toolkits and what is your PXI system (controller and boards)?

Regards,

Wagner Marinho

wagner.marinho@ni.com

National Instruments

Applications Engineering

www.ni.com/suporte

[Carlos A]

Hello.

We solved the problem using the same clock source for the transmitter and the receiver.

Now we need to separate the two modules in different PXI systems with no wire conection, but we would have the same problems again.

Can we solve this issue by using high stability rubidium standards or a GPS synchronization system as clock sources?

Thanks for your answer.

Carlos A. Viteri

Universidad de los Andes

caviteri@gmail.com

[wmarinho]

Hi Carlos,

Can you use sync bits in your acquisition to phase align the acquired data?

You can define sync bits for the acquisition that should allow you to acquire your expected data.

Regards,

Wagner Marinho

[Carlos A]

Hello.

I can't use sync bits because the problem is not to acquire the spected data in baseband (lowpass signal) but to synchronize the local oscillator's frequency and phase.

If the 5661 and the 5651 use their own on board clocks (no synchronization), there is a difference between the real carrier frequency and the defined by software (taking into account the achievable system accuracy.The difference between the transmitter's frequency and the receiver's frequency (typically around 1 KHz - 2 KHz at 2.4GHz with the given accuracy) damages the received baseband signal (see the pictures I uploaded above).

This problem disappears when the 5651 and the 5661 are locked to the same clock. The problem is we have to separate the transmitter and the receiver for our research projects, so using the same clock is not an option.

This is why I'm asking for an alternative synchronization system for the 5651 and the 5661.

Thanks for your answer,

Carlos A. Viteri

Universidad de los Andes

[wmarinho]

Hello Carlos,

The VIs that you posted were only using the drivers.

Do you have the Modulation Toolkit?

With modulation toolkit, you can add sync bits to designate the expected data to achieve a phase reference for the data.

Can you try this?

Regards,

Wagner Marinho