Improved methods for investigating boron related defects in 6H-SiC using deep level transient spectroscopy

Author

Soundararajan, Ramraj ; Shaffer, Janna ; Mazzola, Michael S. ; Younan, Nicholas H. ; Saddow, Stephen E.

Author_Institution

Dept. of Electr. & Comput. Eng., Mississippi State Univ., MS, USA

fYear

1997

fDate

4-6 Aug 1997

Firstpage

165

Lastpage

171

Abstract

A new covariance technique is introduced for parameter estimation of DLTS capacitance transients using Singular Value Decomposition and Prony´s method (SVD-Prony). The SVD-Prony method is shown to be effective at Signal-to-Noise Ratio´s (SNR´s) above 6 dB. The algorithm is shown to accurately estimate the time constants associated with non single-exponential capacitance transients. This paper describes the software written to implement the algorithm using a multi-platform MATLAB Graphic User Interface (GUI). The software also has a wavelet preprocessing tool for SNR´s less than 6 dB

Keywords

boron; capacitance; covariance matrices; crystal defects; deep level transient spectroscopy; graphical user interfaces; parameter estimation; semiconductor materials; silicon compounds; singular value decomposition; spectroscopy computing; 6H-SiC; B related defects; DLTS capacitance transients; Prony method; SNR; SVD-Prony method; SiC:B; algorithm; covariance technique; deep level transient spectroscopy; graphic user interface; multi-platform MATLAB GUI; nonsingle-exponential capacitance transients; parameter estimation; signal-to-noise ratio; singular value decomposition; software implementation; time constant; wavelet preprocessing tool; Silicon compounds;

fLanguage

English

Publisher

ieee

Conference_Titel

High Speed Semiconductor Devices and Circuits, 1997. Proceedings., 1997 IEEE/Cornell Conference on Advanced Concepts in

Conference_Location

Ithaca, NY

ISSN

1079-4700

Print_ISBN

0-7803-3970-3

Type

conf

DOI

10.1109/CORNEL.1997.649355

Filename

649355