spectrum of IQ data

[yy2822205]

I'm using RFSA ni5660 to perform digital downconversion of BPSK signal(from RFSG ni 5670, 2M bandwidth, carrier frequency 14Mhz), Since the DDC output is IQ data, how to get its frequency-power in band spectrum gragh ?

[50ohmTerminator]

Hello yy282205,

I think we have an example program (MT RFSA FSK Demod with Power in Band Measurement.vi) that does exactly what you want. I found it at C:\Program Files\National Instruments\LabVIEW 8.0\examples\Modulation\RFSA examples\Modulation and SMT. I am attaching the VI as a .zip file. Keep in mind that you need to have the Spectral Measurements toolkit installed to be able to run this VI.

Cheers
Malay Duggar
NI

[yy2822205]

Hi,

My latest modified vi. and screen shoot of outcome front panels are attached.
1.I'm not sore whether the "signal in " pin should be wired to the ni5660
Read IQ vi. or the Resample vi, which would cause different outcome as
showed in attached screen shoot.
2.I think that if I set 2M as the symbol rate, then I should get 2M
bandwidth spectral, but I got 3M or 4M. What's wrong or how to get the
desired 2M bandwidth as before it's downconverted?
3.How to decide the value of sample per symbol to calculate the desired
sample rate for Resample vi., if I don't use the system parameters VI?
Thanks,

Yang

[50ohmTerminator]

Hi yy2822205,

We received your email yesterday and here is some information from the email and some clarification.
 
1. The reason you are getting a different graph when you wire the Signal In from the Resample.VI is because your dt is still taken from the original signal. Once you take the dt from the resampled signal you will get the same graph in both cases. So resampling the data cuts down the number of samples you have to FFT.
 
2. The bandwidth specified on the second tab in your example is the same as the bandwidth shown on the graph in my tests. If you specify a symbol rateof 2MHz, you will not get a bandwidth of 2MHz. This is expected because of the raised cosine filter function. The raised cosine function multiplies the tx filter parameter alpha + 1 with the symbol rate and this is the real bandwidth.  You can find more information on pulse shaping filters at http://zone.ni.com/devzone/cda/ph/p/id/200
 
3.  You can use the same formula found in the RFSA PSK Demod shipping example which is samples per symbol *symbol rate.

Cheers
Malay Duggar
NI