Title :
A swept parameter technique for statistical circuit simulation
Author :
Brinson, Mike E.
Author_Institution :
Centre for Commun. Technol., London Metropolitan Univ., London, UK
Abstract :
Qucs and QucsStudio open source circuit simulators have a wealth of built in swept data features, including facilities for linear and logarithmic scans of simulation variables and for setting component values and device parameters. These simulators also allow semicolon separated lists of numerical values to be used as swept data. This little known feature provides a very flexible mechanism for generating component and device parameter statistical data. An outline of a statistical circuit simulation technique is presented in this paper. The proposed technique can be used with any general purpose circuit simulator equipped with swept data capabilities and as such is suitable for the study of device and circuit performance resulting from variations in device parameters and component values. The operation of the proposed simulation technique is illustrated with the results from an investigation of the statistical performance of a simple MOS current mirror integrated circuit cell, modeled with a speed optimized Verilog-A version of a long channel EPFL-EKV v2.6 MOS transistor model.
Keywords :
circuit simulation; hardware description languages; semiconductor device models; statistical analysis; MOS current mirror integrated circuit cell; Qucs open source circuit simulator; QucsStudio open source circuit simulator; long channel EPFL-EKV v2.6 MOS transistor model; speed optimized Verilog-A version; statistical circuit simulation; swept parameter technique; Circuit simulation; Computational modeling; Integrated circuit modeling; Mirrors; Numerical models; Semiconductor device modeling; Transistors; Qucs; QucsStudio; Verilog-A compact semiconductor device models; statistical circuit simulation; swept parameter lists;
Conference_Titel :
Mixed Design of Integrated Circuits and Systems (MIXDES), 2013 Proceedings of the 20th International Conference
Conference_Location :
Gdynia
Print_ISBN :
978-83-63578-00-8