2010 International Electron Devices Meeting 2010
DOI: 10.1109/iedm.2010.5703431
|View full text |Cite
|
Sign up to set email alerts
|

RF CMOS technology scaling in High-k/metal gate era for RF SoC (system-on-chip) applications

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
1
1
1
1

Citation Types

2
30
0

Year Published

2011
2011
2024
2024

Publication Types

Select...
5
3
2

Relationship

0
10

Authors

Journals

citations
Cited by 77 publications
(32 citation statements)
references
References 0 publications
2
30
0
Order By: Relevance
“…As predicted by simulation in Section III, the 50Ω noise figure is 1-2 dB larger than that measured in a 45nm PD-SOI TIA with unitary gain 50Ω output driver [11] suggesting that noise figure no longer improves with scaling from 45nm to 28nm SOI CMOS. This trend has been observed in other bulk and FinFET CMOS technologies [21], [22].…”
Section: Resultsmentioning
confidence: 61%
“…As predicted by simulation in Section III, the 50Ω noise figure is 1-2 dB larger than that measured in a 45nm PD-SOI TIA with unitary gain 50Ω output driver [11] suggesting that noise figure no longer improves with scaling from 45nm to 28nm SOI CMOS. This trend has been observed in other bulk and FinFET CMOS technologies [21], [22].…”
Section: Resultsmentioning
confidence: 61%
“…23(a). This is attributed to the continued scaling of gate oxide [51]. This figure also shows that thick gate devices have higher normalized input referred voltage noise than thin gate devices for all the technology nodes due to the reduced C ox caused by the use of thicker oxide.…”
Section: Phase Noise Analysismentioning
confidence: 75%
“…6[tm,Nr=50). The values of NFmin are higher than for our technology geometry, and the following reasons may be given: (i) NFmin was uncorrected for thermal noise and pad parasitic contributions [3], (ii) NFmin is negatively impacted by the huge number of gate fingers (Nr=300) which contributes to greatly decrease fmaxo through an increase of parasitic fringe capacitances [12] (see also Fig.3 in [13], the variations of fmax as a function of W tot, for different W u and N f as parameter). NFmin is also reported in Fig.6.B for the same technology as the one described in [3], but corresponding to a previous node (45-nm) [12] (and probably a RF transistor better optimized); in this case, our extracted NFmin data are in line with those reported.…”
Section: A Rf Ssec Extractionmentioning
confidence: 97%