We need to demodulate a Continuous Phase Fast FSK (CPFSK) using the NI Lavbiew and RFSA PXI 5660.

[Sunil Rajanna]

Hi Andy,

I have done more checks on the vi that you sent and that which was developed yesterday, since these are closer to what we are looking for. Your vi was run as it is and the spectrum checked, we see that the deviation is not at the required 1.2 and 1.8 Khz form the carrier of 10Mhz instead it is at 300Hz, the same was observed with the vi that i have posted earlier, when observed using an external oscilloscope too we observe the same, this exercise was done to alleviate doubts of sampling and such similar parameter setting errors which may cause aliasing.

I was wordering about a second method of doing the upconversion since I am stuck with that for now.

Since we have a way to generated the continuous phase FM signal manually (see attached vi.) we no longer need to modulate our message signal, the manually generated signal is at a freq of 1.2Khz and 1.8Khz according to the message bits and they are seemlessly connected together,

The method:

We separately generate this signal using the ARB PXI-5421 and us the upconverter PXI-5610 separately

to just upconvert the baseband to the required carrier.

I have started to work on this parallelly, am now at the stage of generating the FM from the ARB independently, there are two waveforms, one of 64 points/bit the other 200 points/bit,the waveforms were checked and seems ok. NOw the need is to upconvert this waveform to the req carrier without further manipulation on this data, say, by just setting the upconverter to 100MHz, and take out the RF signal from the PXI 5610. Can this be done as simply as that or are there any more loop holes!

Pls revert back on this too.

thanks and regards,

sunil

Message Edited by Sunil Rajanna on 07-07-2005 06:02 AM

Message Edited by Sunil Rajanna on 07-07-2005 06:02 AM

[Andy_Hinde]

Hello Sunil,
I think we should take a step back as I believe this is becoming overly complicated. I have found that you can use the Modulation Toolkit Apply Impairments routine to add a complex IQ shift to your signal. I am attaching a ZIP file with several screenshots and a TX and RX application to transmit and receive your signal.

The transmit VI generates a 288 bit pattern of user choice and BFSK modulates it. Then, the Apply Impairments routine shifts it to a baseband center frequency of 1500 Hz (this is configurable on the front panel). Finally, this waveform is written to the PXI-567x hardware via the NI-RFSG driver and generated at a center frequency of 452.9375 MHz. This results in a signal that alternates between 452.939 MHz +/- 300 Hz.

What I would like to stress is that this is exactly the same as if you wrote the program without any IQ shift with the Apply Impairments routine and used a center frequency of 452.939 MHz. If you set the baseband center frequency control to 0 Hz and use a carrier frequency of 452.939 MHz, you will achieve the exact same physical signal output of the 567x. You can see this for yourself with the attached screenshots.

The screenshot Tx With IQ Shift.JPG shows the settings when I use the Apply Impairments to shift the IQ signal. The screenshot Frequency vs Time 1.JPG shows the frequency vs. time data with a center frequency of 452.9375 MHz. The signal alternates between +1200 Hz and +1800 Hz. The associated screenshot Frequency vs Time 2.JPG shows the exact same signal being acquired with a center frequency of 452.939 MHz, and all that has changed is that the signal is displayed as alternating between -300 Hz and +300 Hz. The information is exactly the same.

The screenshot Tx With No IQ Shift.JPG shows the transmit VI with a baseband center frequency of 0 Hz (i.e. no IQ shift) and a center frequency of 452.939 MHz. Now if you look at the receive screenshots Frequency vs Time 3.JPG and Frequency vs Time 4.JPG, you will see that they are exactly the same as Frequency vs Time 1.JPG and Frequency vs Time 2.JPG, respectively.

So my main point here is to explain that you can prove this to yourself with the attached SW and then my recommendation is to remove the Apply Impairments routine and use a center frequency of 452.939 MHz since you will not be using a IQ shift. The receive VI is attached for your convenience in demodulating the digital bits from the acquired waveform. It shows an eye diagram of the signal and a graph of the bits vs. time as well.

Regards,
Andy Hinde
National Instruments

[Sunil Rajanna]

sorry, one of the vis 'generate baseband.vi' was missing from the zip, I have attached the zip once again, please check it out if possible.

I am working on today's iq impairment method of generation sent by you.

thanks,

sunil