O-QPSK Signal generation on 5663E

[chris-a]

What is the best way to generator custom waveform on the NI 5663E sginal generator.  I'm able to generate a QPSK signal with 2 Mchip/s but I can't figure out how to get the offset on Quadrature channel according to bluetooth specifications.  Does anyone know how to do this?

[Dan_K]

Hello Chris,

First, I am assuming that you are using a 5673E for the generation and that you have a 5663E for acquisition.

How are you configuring the waveform for generation? Are you creating the waveform using your own methods or are you using the Modulation Toolkit to create the waveform for generation. If you are using the Modulation Toolkit I would recommend taking a look at the MT RFSG Generate PSK VI, which you can find in the example finder under Toolkits and Modules»Modulation»NI-RFSG.

This example can configure generation of several varieties of phase shift keying, including OQPSK.

Let me know if any of this is unclear or if you have any further questions.

[kantabra]

Hello

I'm making the same and I used the MT RFSG PSK.vi like an example. When I see the spectrum in a spectrum analyzer it is good but if I see it by software (in the graph) is only a baseband signal, so isn't good. I can attach my VI. I think is good to use the FFT Power Spectrum. vi but I don't understand why is different.

Thank you.

[Michelle_N]

I ran the code you have an got this:

I don't think we are using the right method to display the spectrum. I'll see what I can find for you - there should be a better way to get the data you are looking for - seems the hardware is working, just need to find the right VI to show this in your code.

[kantabra]

I'd appreciate to help me because my director doesn't help me anything. He said me that the RF generator modified the signal that I'm generating with labview and I have to model the generator like a VI. I say, I have to do a VI that simulates the funcionality of the generator and then apply it after the signal that I have (after the complex waveform I think) so it should be like a filter and then the signal in software and hardware would be the same.

The problem is that I don't know how to model the generator. I'm working with a RF Signal Generator NI PXI 5652.

Thanks for all.

[Michelle_N]

I think you are probably using a different piece of hardware - the hardware supported by this example are: 5670, 5671, 5672, 5673. Either way, I changed some of the settings on the power spectrum VI and I think the spectrum looks more like what I would expect:

I set dB On to true and restart averaging to true:

Is this what you were looking for?

[kantabra]

No, because this is a baseband spectrum and in the spectrum analyzer I see the signal in the band (i.e 2,4 Ghz the signal appears in the center of that frequency). I think the signal is good, but I don't understand why is center in 0.

[Michelle_N]

Yes, you are correct. Looking at your code should explain why this is. You pass this baseband signal to rfsg_Resample and Write -

This VI takes the baseband signal and includes the I/Q data and writes this new signal out to your card - hence you are seeing two different outputs. You are monitoring a signal prior to the addition of I/Q rate, so that I/Q waveform is created inside the rfsg_Resample and Write VI and sent to your hardware, and then that signal is what is sent out.

[kantabra]

ohhh, maybe is true but I cannont monitoring nothing after that VI, I dont't know how to do it. Only I have 3 outputs and 2 of them are the instrument handle and the error out. The output of the adjusted IQ rate/original IQ rate is 1 because is the same, it doesn't change it. I don't add a new IQ rate, I don't understand it very well...

[Michelle_N]

So, to simulate the upconvert that your hardware is performing, you can use the Upconvert Baseband VI. Note: you need a baseband waveform with enough samples to do the upconversion. For this, you can use the MT Resample VI to resample and give you those extra needed data points. Here is what it would look like:

Give that a try.