Array
(
    [content] => 
    [params] => Array
        (
            [0] => /forum/threads/new-assistant-professor-at-ole-miss-seeking-advice.4308/
        )

    [addOns] => Array
        (
            [DL6/MLTP] => 13
            [Hampel/TimeZoneDebug] => 1000070
            [SV/ChangePostDate] => 2010200
            [SemiWiki/Newsletter] => 1000010
            [SemiWiki/WPMenu] => 1000010
            [SemiWiki/XPressExtend] => 1000010
            [ThemeHouse/XLink] => 1000970
            [ThemeHouse/XPress] => 1010570
            [XF] => 2021770
            [XFI] => 1050270
        )

    [wordpress] => /var/www/html
)

New Assistant Professor at Ole Miss Seeking Advice

GregariousMatt

New member
My name is Matthew Morrison, and I will begin as an Assistant Professor at the University of Mississippi in August. As part of my position, I am building a Computer Engineering graduate program from the ground up, as well as a VLSI System Design Research Lab for CMOS, ASICs, Embedded Systems, Mixed-Signal Systems, and Microprocessors. I want to start a thread to ask for advice on challenges and solutions in building a program. I met Daniel Nenni at the Design Automation Conference, and he recommended I start a thread here.

What are some skills/projects that you would like to see more Bachelors/Master’s students have coming out of school that you believe are not adequately taught or fostered at the university level? I would like to take these ideas and develop them into projects for my students.

On the research end, I believe that Ole Miss has great potential for being a manufacturing and fabless semiconductor research center. They recently built a "Center for Manufacturing Excellence" where they have educational focus on undergraduate interdisciplinary manufacturing education. As you mention at the end of your book, the financial burden of semiconductor design is the a significant challenge, as well as attracting new capital. I see the potential for collaboration in EPSCoR states in developing Fabless labs.

On the academic end, I am looking to develop 5-6 courses. What tools would you recommend for augmenting these courses? What other ideas do you have for building a successful program?

Advanced Digital Systems and Architectures - Organization and design of digital computing systems. Register transfer language; computer architecture; memory; ALU; addressing modes; Parallel Processing; Multicores; Cloud Computing.

Synthesis and Optimization of Digital Circuit Systems - Architectural and logic-level synthesis algorithms , developing a module library in VHDL, coding a synthesis algorithm in C/C++, and using architectural/logic synthesis systems to synthesize designs.

Reconfigurable Computing in FPGAs and Embedded Systems - Introduce internal design of reconfigurable architectures and FPGAs, the CAD tools and algorithms to program these architectures, and application development using these reconfigurable architectures.

CMOS/VLSI Design - Design, layout, simulation, and test of Design custom digital CMOS/VLSI chips, using a CMOS cell library and state-of-the-art CAD tools. Digital CMOS static and dynamic gates, flip flops, CMOS array structures commonly used in digital systems. Top down design example of a bit slice processor.

Design and Analysis of Fault-Tolerant Digital Systems - Basic concepts of dependable computing. Reliability of nonredundant and redundant systems. Dealing with circuit-level defects. Logic-level fault testing and tolerance. Error detection and correction. Diagnosis and reconfiguration for system-level malfunctions. Degradation management. Failure modeling and risk assessment.

Mixed Signal Integrated Circuit Design - An advanced circuits and systems analysis course that computationally efficient manual and computer-aided methods for analyzing the electrical dynamics of both linear and nonlinear models of active networks destined for monolithic realization principally in complementary metal-oxide-semiconductor (CMOS) transistor technologies.

Lastly, I would like any advice or feedback on how you think I could involve SemiWiki in the education process. Do any of you use this forum to augment their courses? If so, what are some pros/cons of such an approach?

Thank you for all of your help and time.

Very Respectfully,
Matthew Morrison
 
1. Never sacrifice academic rigor. Period! Drive students to be excellent, and that while theory never captures everything, understanding theory 100% first is very worth it.
2. Hands-on projects are always useful, but the broader the scope, the better. Touching a lot of pieces of something and never putting them together is never worth it. Fab out isn't as important, but full sims and layout is a great ending point for sense of completeness.
3. Join the Synopsys Academic program and you'll get access to a lot of great curricula, including things like mostly completely projects synthesis through P&R of an entire microprocessor
4. Get students absolutely as SPICE masters, no matter what toolset you use
5. Get on-board with variation analysis in every class, because it matters. Whether its MCMM for digital design and timing budgets, or variation-aware analog design. I'm sure some of the Solido and digital CAD guys around here have strong opinions, but I want people who understand the idea of manufacturing and know how to design for it, whether through fundamentals or through a CAD-specific approach
6. Require some programming skills and associated reasoning skills - or help develop them - data analysis and automation is valuable. Help with setting up toolbenches but if you can have students realize that DUMB automation is bad, but SMART automation is very powerful. ADE GXL is a good example of where you can do both - smart and very very dumb design automation for analog. Same goes for digital toolsets.

Just my 2 cents
 
Frank Wiedmann
Analog IC Design Engineer at Rohde & Schwarz
Take a look at (late) R. David Middlebrook's Design-Oriented Analysis:

R. David Middlebrook: Design-Oriented Analysis Rules and Tools
http://www.rdmiddlebrook.com/D_OA_Rules&Tools/index.asp

Donald Peter: We can do better
http://icee.usm.edu/icee/conferences/asee2007/papers/1362_WE_CAN_DO_BETTER__A_PROVEN__INTUITIVE__E.pdf

Jochen Verbrugghe, Bart Moeneclaey: Integration of the General Network Theorem in ADE and ADE XL
http://analogdesign.be/
 
Hi Matthew,

It was a pleasure meeting you. I too would like more students to get involved with SemiWiki and am open to suggestions. The complaints I hear from the fabless semiconductor companies is that NCGs take too long to train for modern ASIC or SoC design. I hear complaints from Universities that EDA tools are either not available, not configured, or do not include the proper training. Clearly there is a disconnect here which merits further investigation.

Best regards,

D.A.N.
 
Dear professor,

As you correctly noted about financial burden of huge investments, I would highly recommend you introducing a coursework on both macro and micro economics of semiconductor industry to prepare work force for next generation. It would be an asset to have technical folks understand some good economics to work as a team in sustaining progress of fabless semiconductor ecosystem. If you could send me an email at contact@apekmulay.com, I would be more than happy to help you in this regard.
 
Back
Top