What's the minimum NF(Noise Figure) for a CMOS LNA for 2.4 GHz nowadays? (for example using 0.18 um CMOS technology)
Has anybody any idea?
What could the technology nowadays?
I have designed a 1.8 dB NF LNA for WLAN 2.4-2.5 GHz in 0.18 um Si CMOS. The measured NF was around 2.6 dB (indirect measurement).
Some SiGe process based LNA (MAXIM) for 2.4 GHz may have lower NF but thats a trade-off.
Regards
Maddy
Hi.
It depends on what kind of topology you are using. For 2.4Ghz, I have seen NFs around 2.5dB for differential and 1 dB for single-ended. I have designed a differential LNA with 2.8dB NF.
For WLAN where you do not have big blockers the lna can have a lot of gain so the noise figure can be low. At 0.18um cmos a NF of 1.5dB is fesable. The input network Q will be high which can effect linearity as well as component variation sensitivity.
I have designed a 1.8 dB NF LNA for WLAN 2.4-2.5 GHz in 0.18 um Si CMOS. The measured NF was around 2.6 dB (indirect measurement).
Some SiGe process based LNA (MAXIM) for 2.4 GHz may have lower NF but thats a trade-off.
Regards
Maddy
We used the Y-Factor method as the Spectrum Analyzer we used didnt have the Noise Figure personality and the Noise Figure meter. The shift in the noise floor is measured with the Noise Source (connected to the DUT) on and off. Then using a formula, the NF is calculated. This link below gives a brief overview. There is a appnote on NF measurement on Agilent website.