Autocorrelation Analysis as a Technique to Study Physical Mechanisms of MOSFET Low-Frequency Noise

• Published in: IEEE transactions on electron devices Vol. 64; no. 7; pp. 2919 - 2926
• Main Authors: Both, Thiago H., Croon, Jeroen A., Banaszeski da Silva, Mauricio, Tuinhout, Hans P., Scholten, Andries J., Zegers-van Duijnhoven, Adrie, Wirth, Gilson I.
• Format: Journal Article
• Language: English
• Published: New York IEEE 01-07-2017; The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
• Subjects: 1f noise; Autocorrelation; Correlation; Correlation analysis; Low-frequency noise (LFN); MOSFET; Noise; Noise prediction; random telegraph signal (RTS); Semiconductor devices; Silicon; Sociology; Spectra; Superposition (mathematics); Transistors
• ISSN: 0018-9383; 1557-9646
• DOI: 10.1109/TED.2017.2703671

A low-frequency noise (LFN) analysis technique is introduced, based on the autocorrelation of the LFN spectra in terms of frequency, biasing, and temperature. This technique reveals information about the mechanisms behind 1/f noise that is difficult to obtain otherwise. These correlation analyses provide strong evidence that the LFN of both nMOS and pMOS transistors is mainly composed of the superposition of thermally activated random telegraph signals (RTSs). The methodology enables quantification and validation of old and new LFN models, and, therefore, helps to clarify the relation between RTS and 1/f that is still frequently debated in the literature.