|
From: | Richard Jaeger |
Subject: | Re: [Discuss-gnuradio] USRP Gain |
Date: | Fri, 5 Sep 2008 12:18:20 -0500 |
Q3: MDS measurement The signal disappears into the noise floor with the attenuator at 66 dB or 10 mV/2000 = 5 uV input signal. I believe the USRP ADCs are 12 bits with VFS = 2V, so the LSB is 2V/4096 or about 488 uV. How is the system resolving signals that are 20-40 dB below the LSB size? I think I figured this out over night. It's been quite a while since I looked at A/D converter response to very small signals. If an ideal ADC is somehow biased at a code step threshold point, any size signal will cause toggling between two ADC codes, and a subsequent fft will readily pick out the frequency components related to this toggling. Signal size (on the low end) is ultimately limited only by the resolving capability of the comparator element(s) in the converter. If the ADC input sees wide-band noise that is 1 LSB or more in amplitude, the noise acts as a dither signal, and the input no longer needs to be biased at a code step threshold. Adding a very small sinusoid to the noise will alter the ADC code pattern in a detectable way. Again, the signal size should be limited by the resolution of the comparator. Dick... |
[Prev in Thread] | Current Thread | [Next in Thread] |