[moved] RF to ADC or downconverted RF to ADC

Dear All,


I am working on phase array antennas to do null-steering.

I have read lots of papers in this field, but still I have this question in my mind.


Which one is the better approach and why?

We go directly from antenna to ADC at RF (1.5GHz). Or first we do downconversion to IF (200MHZ) and then go to ADC? What are the differences? Converting RF to digital needs more bits?



Thanks for your help in advance.

Assuming that your 1.5GHz ADC and your 200MHz have identical specs (BIG assumption) then the other things to consider are:

Sampling at 1.5GHz will require 7.5 times more data storage (assuming you're storing the samples).
The downconverter is going to add some signal degradation

I don't know what you mean by "Converting RF to digital needs more bits?" More bits than what?

Hi Barry,

Thanks for your help.
By "Converting RF to digital needs more bits?" I meant: For example if we want to digitalize a RF signal at 200MHz each sample to be saved on memory needs 20 bits. Then if we work with the 1.5GHz signal without downconverting, each sample needs more bits or not? I don't know much about digital stuff.


Why is assuming 1.5GHz ADC and 200MHZ ADC have identical specs is a big assumption?


More than signal degradation and storage, is there any other difference?



Regards

You cannot use "direct digitization" since your signal level is too low.( reflected signal level is too low due to weak reflections from the objects to my knowledge )Therefore using direct signal processing will not be efficient.Converting to a lower frequency will you make more flexible to process the signal.

Well, I am supposed to increase level of signal by LNAs after my antenna. When I say "direct" I mean without converting to lower frequencies. Do you think there would be any problem if level of signal is enough? Also, how much power the signal must have to say the level of signal is enough?

Thanks

I prefer process this signal at lower frequencies ( @IF ) to obtain better result.Because processing a signal at lower frequencies will need lower sampling frequency,less bit storage,less complexity.
Also, filtering unwanted jamming effect can be tolerated easily by filtering somewhere in the middleway.

Do to TI's or Analog's website and compare essential parameters for ADCs running at 200 Msps and 1.5 Gsps.

Your question is a little meaningless until you explain your exact end use, AND your criteria for "better".

Does "better" mean :1)cheapest, or 2) best null depth, and so on

if cost is no object, go find a high bit 4 ghz clock rate adc and be happy