Passband sampling or DDC?

[agump]

passband sampling

Hi,
I am engaged in a project which need to process digital IF signal. The passband sampling theory shows that when we sample a signal based on the theory, we can get the signal in the base band. But when I read some references , they show me that the digtal down converter prevail against the passband sampling. I am confused about how to select which one to carry out my task . Why not use the passband sampling to convert the signal to baseband , since passband sampling need much less resource and efforts .Please give me some help. Thanks.
Best regards.

 

[fontp]

ddc sampling jitter

The reason is - sampling jitter. You have to have timing generator of perfect quality and proper design

Read AnalogDevices Apllication Note AN-501
"Aperture Uncertainty and ADC System Performance Application Note" (AN-501)
http://search.analog.com/search/default.aspx?query=AN+-+501&local=en

 

[agump]

sampling passband signals

Thanks for your reply. I have read this article and can't understand why the aperture jitter make the bandpass sampling unusable , since the aperture jitter is determined mainly by the ADC and the sampling clock. If using the bandpass sampling , a low-speed clock can be used. In my option , I think this can reduce aperture jitter.

 

[fontp]

bandpass sampling jitter

But jitter noise figure is propotional to analog Fif. So jitter limits direct sampling for high Fif. Another help from DDC is processing gain - we can use relatively low-bit-number ADC

 

[agump]

ddc processing gain

What do you mean about "jitter noise figure is propotional to analog Fif", I think bandpass sampling and DDC both process the same Fif and just select different sampling frequency. I am not family about the processing gain and I am about to read some reference about it. Thanks.

 

[fontp]

sampling passband

You are right, this has no direct respects to your question. This issue is from other side - Why do we use analog mixer, while simple undersampling works like mixer as well? When sampling transforms Fif to DC as an alias. So simpliest receiver is - ADC alone, without mixers and DDC. But jitter and noise...

Processing gain is the property of DDC - we spread ADC quantization noise around wide spectrum range (using high-speed ADC) and then supress this noise by filtering.