A. Matthew Francis, Ph.D. (IEEE M’00) received B.S. degrees in electrical engineering and physics (2003, 2004), the M.S. degree in electrical engineering (2007), and Ph.D. degree in electrical engineering (2009) from the University of Arkansas under Dr. Alan Mantooth. In 2011 he founded Ozark IC, Inc., a provider of extreme environments IC solutions. Research interests include simulation and modeling, compact modeling of radiation effects on semiconductors, design flow methodologies and algorithms for analog and mixed-signal modeling.His industry experience as a student included stints in VLSI/ASIC modeling and verification with Hewlett-Packard, Inc. and mixed-signal CAD research work with the Air Force Research Lab at Wright Patterson AFB, OH. In addition, as a undergraduate and later graduate researcher he led the University of Arkansas Mixed-Signal CAD lab's Paragon Modeling Environment effort and CAD tools group. From 2005-2010 he was a founding member and senior applications consultant for Lynguent, Inc, in Fayetteville, AR, the company which commercialized the Paragon analog modeling IDE as the ModLyng Integrated Modeling Environment (IME). He has served in several capacities for Tau Beta Pi, IEEE and Eta Kappa Nu student organizations.
Links
Current and Past Projects
"High Fidelity Compact Models for SiGe BiCMOS Charge-Sharing Single Event Analysis": (Author, BSIM3 CMX model with radiation support)
This work is funded by NASA and involves creation of "white-box" compact models to support the IBM Silicon-Germanium (SiGe) BiCMOS 5AM process. This includes creation of ModLyng spectre-compatible BSIM3v330 and Mextram wide-temperature compact models validated over a large variety of temperatures and biases. Additional work will allow insertion of characterized charge-sharing effects directly into the compact model topologies through use of ModLyng CMX technologies [15].
"High Fidelity Compact Models for 45nm SOI Charge-Sharing Single Event Analysis": (Author, BSIM4 SOI CMX model with radiation support)
This work is funded by the Defense Threats Reduction Agency (DTRA) and involves creation of "White-box" compact models to support the IBM Silicon-On-Insulator (SOI) 45nm process. This includes creation of ModLyng spectre-compatible BSIMSOI41 models validated over a wide range of temperature and biases. Additional work will allow insertion of characterized charge-sharing effects directly into the compact model topologies through use of ModLyng CMX technologies.
"Efficient Charge-Sharing Single Event Effects in Compact Models": (Author/Creator: BSIM4 CMX model, LynRad automated radiation analysis tool)
This work is funded by the Defense Threat Reduction Agency (DTRA) and involves a modeling method to solve many of the problems faced in simulating the effects of radiation on semiconductors in SPICE-like simulators. As part of the research an extremely high-fidelity open (verilog-a compatible) BSIM4 model was developed using CMX and designed to be extensible for any class of new effects ranging from radiation to thermal and variability effects. [1-2], [13-14]
"Common Modeling Language (eXchange)". (Co-Author, CMX Format)
This work initially funded by the Semiconductor Research Corporation (SRC) led to the development of the Common Model eXchange format (CMX). CMX is an XML-based description medium for exchanging mathematical models of complete IC's, blocks, sub-systems and components which grew out of the Paragon project (below). It is a superset of all commonly used Hardware Description Languages (HDLs) including Verilog and VHDL and serves as the storage medium for the ModLyng tool already mentioned.
"Paragon Mixed-Signal Modeling Tool": (Co-creator/desginer, Paragon Modeling Tool)
This work at the University of Arkansas MSCAD lab was funded in part by DARPA, NSF and ONR in involved the creation of the first integrated modeling tool for Hardward-Description-Language (HDL) models. The tool was an early adopter of the XML format in EDA, as utililzed for storage of an abstract representation of a HDL model, independent of specific HDLs. Using this method a user could create one model and export to any supported language, e.g. Verilog-AMS, VHDL-AMS or MAST. This work has been licensed and commercialized by Lynguent, Inc. as the ModLyng product. [3] [6] [9-12]
"Ascend: Bottom-up Behavioral Model Creation Tool": (Software Development Supporter)
This work at the University of Arkansas MSCAD lab was funded in part by DARPA and involved the create of a modeling tool for the automatic creation of behavioral (MAST or VHDL-AMS) models of analog circuits. For a wide variety of analog applications (comparators, opamps, etc) the tool automatically generates an equivalent HDL model from a circuit netlist which may run upwards of 1000x faster than a spice-primitive circuit design. The author's contribution involved support of the software design and GUI development of the Ascend tool. [2]
mailto:matt@amfrancis.com
Refereed conference proceedings and journal articles of IEEE, IEE and other recognized conferences and publications.
(A.M. Francis, A. Francis or M. Francis.)
Recently accepted:
[1] A. M. Francis (Member IEEE), N. W. van Vonno (Senior Member, IEEE), J. Holmes (Member IEEE), "Predictive Modeling of TID and SET Effects from Datasheet and Radiation Data for Commercial Components", Accepted for presentation at the 2011 Nuclear and Space Radiation Effects Conference (NSREC), Las Vegas, NV, July 24th, 2011.
Recently Published
[13]
[14]
[15]
A.M. Francis, D. Dimitrov, J. Kauppila, A. Sternberg, M. Alles, J. Holmes, and H.A. Mantooth, “Significance of Strike Model in Circuit-Level Prediction of Charge Sharing Upsets,” Nuclear Science, IEEE Transactions on, vol. 56, 2009, pp. 3109-3114.
J.S. Kauppila, A.L. Sternberg, M.L. Alles, A.M. Francis, J. Holmes, O.A. Amusan, and L.W. Massengill, “A Bias-Dependent Single-Event Compact Model Implemented Into BSIM4 and a 90 nm CMOS Process Design Kit,” Nuclear Science, IEEE Transactions on, vol. 56, 2009, pp. 3152-3157.
C. Webber, J. Holmes, A.M. Francis, R. Berger, A. Mantooth, A. Arthors, K. Cornett, J.D. Cressler, “Event driven mixed signal modeling techniques for System-in-Package functional verification,” in Aerospace Conference, 2010 IEEE, pp. 1-16, 2010.
Published