The Designer's Guide Community Forum

The Designer's Guide Community Forum
https://designers-guide.org/forum/YaBB.pl
Design >> RF Design >> Cascaded Noise Figure (SSB / DSB)
https://designers-guide.org/forum/YaBB.pl?num=1095149658

Message started by ThomasLee on Sep 14^th, 2004, 1:14am

Title: Cascaded Noise Figure (SSB / DSB)
Post by ThomasLee on Sep 14^th, 2004, 1:14am

Hello everybody!

I don´t know if this has been discussed before, but I hope you can help me.
I have a general problem with system calculation.
Think of a system with these blocks: LNA with 2dB Noise Figure and 15dB available gain, a mixer with 12 dB Noise Figure (SSB) and e.g. 5 dB gain.
These blocks shall be used for a direct conversion receiver. Here my assumptions, correct me if I am wrong:
For a direct conversion receiver the DSB noise figure is applicable because there is no image. So I subtract 3 dB from 12 dB and I get 9 dB DSB NF for the mixer. With the LNA gain and NF I can calculate the cascaded NF to 2.56 dB.
But if I don´t subtract the 3dB first (like for SSB), I use 12 dB for the cascaded NF calculation, then I get 3.13 dB total NF.
So far so good. Then I say take this whole block, regard it as a low noise mixer and use it for direct conversion, so I can subtract 3dB and I´d get a system with 0.13 dB noise figure!!! What am I doing wrong?

Thank you!

Title: Re: Cascaded Noise Figure (SSB / DSB)
Post by Eugene on Sep 14^th, 2004, 8:43pm

Let the signal bandwith at one of the ADCs be "BW". The signal bandwidth at the receiver input is 2*BW. The cascaded noise figure assumes the input noise density is 4kTR. 4kTR is a single-sided density spread over 2*BW on the RF side of the mixers. However, when you derive the cascade formula and compute the contribution from the baseband legs, you must acknowlegde that the single-sided density of 4kTR is only spread over BW, not 2*BW. The result is a reduction of ONLY the baseband noise by a factor of 2 in the cascade formula, not the noise in the entire formula.