MICROELECTRONICS SYSTEM-DESIGN LAB(MSDL)
Research work at The Microelectronics Systems Design Lab (MSDL) is currently focused on:
"NEUROMORPHIC INTEGRATED SYSTEMS FOR ON-SITE and ON-TIME DECISIONS APPLICATIONS." and
MICROELECTRONICS SYSTEM-DESIGN LAB(MSDL)
NEUROMORPHIC AND BIO-INSPIRED TECHNOLOGIES FOR DECISION AND
ACTUATION APPLICATIONS (MSDL=old name)
Our work on NEUROMORPHICs include, but not limited to:
We are one of few Universities dealing with The New Neuromorphic approaches.
PARTICULARLY INTEGRATED AND IMPLEMENTED SYSTEMS for Hardware On-site and on-time decisions making.
SAMPLE PROJECTS
Winter
2009, "Sampling Spiking Neural Network" by J. Allen
Winter
2009, "16-Bit Floating Point Multiplier," by David VanSickle, ECE585
Winter
2009, "Nega-binary Code Converter, Multiplier, and Counter using RHDL" by J. Allen, 2008-2009
Winter
2009, "A Reconfigurable Spiking Neural Network Digital ASIC," By Kevin VanSickle, ECE585-485
xxxThe Microelectronics System Design Lab at Oakland University (MSDL)
is recognized research center for design, testing and evaluation of prototype
Very Large Scale Integrated Ciruits (VLSIC).
The MSDL has the expertise, tools and facilities needed to build advanced
microelectronic systems for the neighboring industrials. These include:
Application Specific Integrated Circuits (ASICs); Integrated signal processing
circuitry; Sensors and actuators integrated interfacing circuits; Neural
Network; Analog and Digital Implementation for smart ICs; Automotive Electronics
Design, Evaluation, Implementation, and Testing; also using Residue Number
Systems algorithms and prototypes for efficient computing and error detection
and correction.
Until the beginning of the nineties, ASICS had only two options for
its implementation: (1) Design for optimum performance by using the expensive,
time consuming, more prone to errors FULL-CUSTOME DESIGN. (2) Design
using available sets of libraries or gates for short time-to-market, and
a reduced error rate SEMI-CUSTOM DESIGNS.
Currently, microelectronic applications require a mix of full-custom
and semi-custom designed components, as well as a hybrid of analog and
digital elements. These can be implemented in monolithic integrated circuits,
or the more practical Multi-Chip-Modules (MCM), Flat-Panel-Display technologies
(FPD), or Micro-Electro-Mechanical-Systems(MEMS).
The effective use of Electronic Design Automation (EDA) tools represent
a bottleneck in fully exploiting microelectronic systems. Computer H/W
and S/W tools are available at the MSDL to design, implement and evaluate
effective microelectronics systems. Adequate electronic testing facilities
are available. Chips are processed through the ARPA/NSF supported MOS Implementation
Services (MOSIS).
Similarly extensive simulation tools are available for accurately predicting
the behavior of a design that exists only as a file. Testing a design file
by simulation may often be enough to ascertain the behavior the design.
However, when dealing with microsystems actual hardware must be
built, tested and evaluated.
The achievement of smart sensors and actuators, particularly within
the automotive industry, could revolutionize the accuracy of many engineering
systems as well as allow the performance of jobs not now feasible.
Within the past few years, efforts in the smart sensors and actuators
area are growing rapidly nationally and worldwide with major focus on:
Microsensors, Microactuators, and appropriate interface circuits all integrated
monolithically or hybrid modules using: MEMS, MCMs, or FPDs.
The MSDL supports graduate and undergraduate courses as well as basic
and applied research. The MSDL activities have resulted in a number of
Ph.D. and M.Sc. dissertations, over 130 student projects, and numerous
publications in conferences and journals.
The lab is Supported by funds from OU Research, Michigan Research Excellence
Fund, NSF, NSF/SBIR, ARPA, US-AIDS, OU Educational Development, and ARPA/MOSIS.
RESEARCH ACTIVITIES
Smart Integrated Automotive Microsystems (SIAM); Neural Networks for
Gas Sensors; Integrated Circuit Verification; Multi-Chip-Modules Alternative
Technologies; Integrated Sensors and Actuators Circuit Interface; Efficient
Computation Using RNS; RNS Error Detection and Correction; Interconnects
at the wafer-level (SWAN) and Chip-Level; Color Space Conversion for Manufacturing
Inspection and Robotics Precision; Analytical and Numeral Semiconductor
Device Models; Power Management for Sensor Clusters; Neural Networks for
Signal Processing; Recurrent Dynamic Neural Networks (RDNN) and Self-Organizing
Feature Maps (SOFM); Micro-Electronic-Mechanical Systems (MEMS).
RELATED COURSES
Undergraduate courses
Graduate courses
DESIGN AND SIMULATION TOOLS
MAGIC, OCT (VEM), and L-EDIT: VLSIC Layout and physical designs
EE585/485 VLSIC Design of Digital ASICS MOS integrated circuit fundamentals.
NMOS, PMOS and CMOS processes. Inverters, logic and transmission gates,
propagations delays. Design techniques, performance evaluation, speed/area/pwer
trade off, finite state machines, memory and registers. VLSI Systems design
projects for semicustom integrated circuits using standard cell libraries
and /or PLAs within the OCTOLLLS. Four Credits.
EE683 AdvancedVLSIC Analog/Digital System Design Full Custom design,
analysis and techniques of ASICs. Metal-Oxide-Semiconductor devices, circuits
and future trends, MOS processing and design rules. Analog and Digital
VLSI MOS circuits, design techniques and performance evaluation. Integrated
system fabrication and the large scale integration design route. Data and
control flow in systematic structures. Implementing integrated geometry
to wafer fabrication. Advanced analog/digital systems design. Four credits.
Projects involve semi-custom design use a top-down approach with a High-Level
Description Language (HDL), followed by simulations at the functional-level,
the logic-level, and the swtch-level. Other projects with, complete or
partial, full-custom designs use a bottom-up approach with physical layout
of the design using MAGIC, VEM, or L-EDIT followed by parameters extraction,
device and then circuit level simulations. All the projects consider design
for testability. Testing, evaluation and validation of processed prototype-chip
is performed.
COMPLETED AND TESTED CHIP DESIGN
Speed Proportional Steering IC. Analog Adder and Sensor Power Management.
ASIC for Color Space Conversion. Squaring Function, Median Circuit, and
Clock Generators. New Residue Based 7-bits Arithmetic Divider. Residue
to Mixed-Radix Decoder and Sign Detector. Efficient Arithmetic Logic Unit
using RNS. Power Management for Environmental Sensor Cluster. 16-Input
Max/Min Analog Circuit Selector for Gas Detection. Lighting Control Module
ASIC. Self-Organizing Feature Map (SOFM).
1. WE are the FIRST to CONCEIVE, DESIGN and IMPLEMENT The COMPLETE Decision Making
Spiking Neural Network Electronic-nose.
Our Integrated system approach to SNNs mimic that of the biological spiking phenomenon
for potentiating or depressing biological neurons synaptic connections (as measured in monkey's brain).
Using the unique capabilities of our SPIKING NEURAL NETWORKS (SNNs) many other applications can benefit
such as Radar Communications, signal processing and de-noising, all biological sensors,
detectors, and decision making applications.
2. Polymer Electrolyte Transistors (PET) for Novel Hybrid devices for new dimensions
in biological/medical applications and polymorphic computing.
3. Graphite Piezo-Electro-Chemical Transducer (PECT) effect and applications.
We are in the process of making graphite/graphine Quantum-switches, and quantum impedance devices.
These promise to have great applications in Tera Hertz detection applications, Safe Diagnostics-
rather than the harmful X-rays, In-vivo Medical sensors and actuators, as well as power-harvesting applications.
4. MULTI-FUNCTIONAL LOGIC Circuits and Systems using biological and amorphous materials.
These facilitate for Energy Harvesting Systems, and Totally Reconfigurable Analog (better compatible
with real-life applications) systems.
EE384 Electronic Material and Devices
EE378 Design of Digital Systems
SYS 431 Automatic Control Systems
SYS 433 Modern Control System Design
EE485 MOS VLSI, Semicustom Design of Digital ASICs.
EE487 Integrated Electronics
EE495 Special Topics
EE490 Senior Project
EE581 Integrated Circuits and Devices
EE585 MOS VLSI Semicustom Design of Digital ASICs
EE683 Advanced MOS VLSI Analog/Digital Systems Design
EE595 Semicustom Design of ASICs using Field Programmable Gate Arrays
EE595 Special Topics
EE690, EE691, EE795 Graduate Projects and Thesis
EE572 Microcomputer-based Control Systems
SYS563 Foundation of CAD
SYS569 Computer Simulation in Engineering
EE637 Digital Signal Processing
EE638 Digital Image Processing
EE682 Field Effect Devices
EE485/585 and EE683 along with special topics senior and graduate microelectronics
research projects mostly rely on the MSDL where chip designs and implementations
are carried out and then processed through the MOSIS facilities.
SUPREME: IC process simulator
PISCES and MINIMOS : Device level simulators
SPICES3E and HOSPICE: Circuit level simulators
OCTTOOLS, OASIS, and TANNER TOOLS: Integrated VLS electronic systems
design suites
XILINX and VIEWLOGIC.
SAMPLE PROJECTS
Winter 2009, "Sampling Spiking Neural Network" by J. Allen
Winter 2009, "16-Bit Floating Point Multiplier," by David VanSickle, ECE585
Winter 2009, "Nega-binary Code Converter, Multiplier, and Counter using RHDL" by J. Allen, 2008-2009
Winter 2009, "A Reconfigurable Spiking Neural Network Digital ASIC," By Kevin VanSickle, ECE585-485
Winter
1998, "Roulette Game on a Chip" by S. Shaik
Winter
1998, "Programable Digital Sigmoid Function, Digital Sinusoidal Oscillator,"
by A-Alkhateer and M. Al-Nsour
Winter
1998, "Automotive Seat Heat Interface Module (SHIM) ASIC," by M. Wheaton.
Winter
1997, "ASIC for Liquid Level and Temperature Measurement Controls Sequence,"
by J. Begin and D. Missar
Winter
1997, "Single IC Vehicle Body Control Module," by S Makar and K. Grand
Fall
1996, "OP-AMP., and SuperPMOS for Piece Wise Linear programmable
Function Generation," by J. Porcell
Fall
1996, "A Full Integrated Audio-Frequency Sinusoidal Oscillator," by E. Hoekstra
Fall
1996, "Neural Network Analog Building Block: 4 Quadrant Multiplier and
Sigmoid Function," by M. Nsour
1995,
"Residue to Mixd-Radix module conversion and sign Detection,"
1995,
"Novel Capacitor Design for Static Clock Generator: Poly 1/Poly 2, and
Poly/n-n+ Diffusion, Squaring Function"
1995,
"Self-Organizing Feature Mapping NN"
1995,
"Designing and VLSI Implementation of an ASIC for Color Space Conversion"
by M. Nsour