Measurement and extraction of RTS Parameters under 'Switched Biased' conditions in MOSFETS


Share/Save/Bookmark

Kolhatkar, J.S. and Wel, A.P. van der and Klumperink, E.A.M. and Salm, C. and Nauta, B. and Wallinga, H. (2003) Measurement and extraction of RTS Parameters under 'Switched Biased' conditions in MOSFETS. In: 17th International Conference on Noise and Fluctuations, ICNF 2003, 18-22 August 2003, Prague, Czech Republic (pp. pp. 237-240).

[img]
Preview
PDF - Published Version
1MB
Abstract:With decreasing device dimensions, the low-frequency noise power spectrum of MOSFETs is dominated by Lorentzians, arising from Random Telegraph Signals (RTS). Previously, it was shown that the low-frequency noise of MOSFETs decreases, if the transistors are switched “off” periodically (switched bias conditions). This low-frequency decrease in noise can be modeled qualitatively by instantaneously changing the RTS time constants using a cyclostationary numerical RTS model. However, until now all ‘switched biased’ measurements were in the frequency domain. In order to validate the assumptions made in the model, it is highly desired to measure and extract the RTS time constants, and amplitude from the time domain measurements for constant and ‘switched biased’ conditions. In this paper, we present a new RTS parameter extraction procedure in the time domain by which it is possible to measure; the mean capture and emission times (τc, τe), and the amplitude, under ‘switched biased’ conditions. We also present measurement and RTS parameter extraction results on a PMOS device (W/L=10/0.3). The extracted RTS parameters are then fitted in an analytical noise power expression, and compared with an RTS noise power measurement using a spectrum analyzer.
Item Type:Conference or Workshop Item
Faculty:
Electrical Engineering, Mathematics and Computer Science (EEMCS)
Research Group:
Link to this item:http://purl.utwente.nl/publications/45795
Export this item as:BibTeX
EndNote
HTML Citation
Reference Manager

 

Repository Staff Only: item control page

Metis ID: 213272