Abstract
A deep understanding of how to reduce flicker phase noise (PN) in oscillators is critical in supporting ultra-low PN frequency generation for the advanced communications and other emerging high-speed applications. Unfortunately, the current literature is either full of conflicting theories and ambiguities or too complex in mathematics, hiding the physical insights. In this brief, we comprehensively review the evolution of flicker noise upconversion theories and clarify their controversial and confusing parts. Two classes of such upconversion mechanisms in voltage-biased LC-tank oscillators (nMOS-only and complementary) are specifically compared and numerically verified using a commercial simulation model of 28-nm CMOS. We identify that non-resistive terminations of both 2nd and 3rd harmonic currents contribute to oscillation waveform asymmetries that lead to the flicker noise upconversion. Further, we discuss three 1/f <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">3</sup> PN reduction mechanisms: waveform shaping, narrowing of conduction angle, and gate-drain phase shift.
Highlights
A S THE CMOS technology advances, the flicker (1/f ) noise upconversion in oscillators has become a serious issue due to the increasingly worsening intrinsic 1/f noise in MOS transistors, especially in FinFETs [1] and FDSOI [2]
The emerging applications of 5G/6G wireless communications [3]–[13], high-speed wireline links [14], [15], and quantum computing [16]–[18] all rely on frequency sources of ultra-low phase noise (PN)
In contrast to “symmetrising” the oscillating waveform by VH2 [48], narrowing of conduction angle could decrease the flicker current noise exposure to the asymmetric waveform caused by the 2nd or 3rd harmonic current entering the non-resistive terminations
Summary
A S THE CMOS technology advances, the flicker (1/f ) noise upconversion in oscillators has become a serious issue due to the increasingly worsening intrinsic 1/f noise in MOS transistors, especially in FinFETs [1] and FDSOI [2]. Two types of 1/f 3 PN mechanisms dominate: 1) frequency perturbation due to the large, sensitive, tuning-tank varators, Cvar, and non-linear parasitic capacitance of connected devices (e.g., M1/2); 2) instantaneous phase perturbation due to an injection of flicker current noise into the tank directly from the cross-coupled pair, M1/2, or indirectly from the tail current source, Mtail, via M1/2 The former mechanism of AM-FM by Cvar was well explained in [33]–[35]: both differential-mode (DM) and common-mode (CM) oscillation waveform amplitudes depend on the tailcurrent and are slowly modulated by its 1/f current noise. The AM-FM mechanism has been minimized with the scaling of CMOS technology and the introduction of switched-capacitors (Csw-cap) in both VCOs and digitally controlled oscillators (DCOs) [46], [47] As a result, this leaves the cross-coupled pair as the only remaining source of the 1/f noise upconversion. We discuss comprehensively the 1/f noise upconversion mechanisms from the cross-coupled pair as well as several 1/f 3 PN reduction methods using a unified theory framework [47]– [51]
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callMore From: IEEE Transactions on Circuits and Systems II: Express Briefs
Disclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
