The
University Senate of Michigan Technological University
Proposal
6-09
(Voting Units: Academic)
“Proposal for a Doctor of Philosophy Degree Program
In Computer
Engineering”
1
Introduction
1.1
General
Description
This is a proposal to establish a Doctor
of Philosophy (PhD) program in Computer Engineering (CpE) within Michigan Technological
University’s Department of Electrical and Computer Engineering (ECE). The
mission of the proposed graduate program is to educate engineers in the science
and technology of this field and to recognize their achievement by creating an
advanced Computer Engineering degree at Michigan Tech. Graduates of the program
will have the necessary skills and will be highly qualified to perform
scientific and technologically advanced research to solve problems in the
design, development, and implementation of complete computer-based systems and
application domains.
1.2
Rationale
Computer Engineering (CpE) is a hybrid
discipline, born of two parent disciplines, Computer Science (CS) and Electrical
Engineering (EE). It has, for decades, been recognized by the Accreditation
Board for Engineering and Technology (ABET) as a separate discipline, with
academic content distinct from both EE and CS [1].
1.2.1 National Trends
As of October 2006, there were 188
ABET-accredited undergraduate CpE degree programs in the United States [2]. In
addition, between 1996 and 2006, the number of programs in Computer Engineering
and Technology increased 111%, making it the
fastest growing single discipline
reported by ABET [2].
Nationally, of the 252 universities with
graduate engineering programs, the overwhelming majority offer Computer
Engineering graduate degrees in one form or another. Most of these programs
bundle CpE with either CS or EE under an umbrella title, such as Computer Science and Engineering, or Electrical and Computer Engineering.
However, there are now 40 standalone doctoral degree programs in Computer
Engineering [3]. In Michigan, the
University of Michigan-Dearborn, Wayne State University and Western Michigan
University offer the PhD. in computer engineering. Only two PhD degrees in
computer engineering were granted in Michigan in 2006-2007; 193 Ph.D. degrees
in computer engineering were granted nationally in that year.
1.2.2 Related Graduate Programs
Michigan Tech currently offers PhD
degrees in CS, EE, and Computational
Science and Engineering (CS&E). Graduate study in computer engineering
is now included within the EE degree programs.
This mix of computing disciplines is
typical of programs at other universities. These inter-related disciplines can
be defined by where their particular focus area lies within the broad spectrum
of computing topics. Although there can be considerable overlap between the
different computing fields, the differences
are best described relative to the “center of mass” of each field. For example:
·
Computer
Science traditionally focuses on the
theoretical and software aspects of the process
of computing, rather than on the computer system itself. While some
computer scientists do delve into computer architecture, the center of mass is
toward hardware-independent topics, with little emphasis on hardware
architectures, and even less interest in electronic circuitry.
·
Electrical
Engineering traditionally represents
the opposite extreme, focusing on the physics, electronics, circuitry, and
related aspects of both analog and digital hardware, with little or no emphasis
on software design, hardware/software integration, or computational theory.
·
Computational
Science and Engineering focuses on
the application and use of high
performance computing platforms to solve scientific and engineering problems.
In this field, the computer is simply a tool used to achieve some research goal
in an unrelated scientific or engineering field; thus, the computer system is merely
a means to an end, not the end in
itself.
·
Computer
Engineering treats a computer-based
system as one spanning a continuum of technologies on both sides of the
traditional analog/digital systems
boundary as well as the traditional hardware/software
boundary. Thus, a computer
engineer studies the whole computer system in its entirety, is equally comfortable working with
both hardware and software, and has an intimate understanding of how the
hardware and software interact with each other. S/he can thus integrate all of
these technologies into a single system, write hardware-dependent software,
evaluate hardware/software trade-offs, and engage in hardware/software co-design. These abilities make the Computer
Engineer uniquely qualified to conceive, design, and build complete
computer-based systems to serve a wide variety of applications.
1.3
Projected Enrollment
While exact enrollment numbers are
difficult to predict, estimates can be extrapolated from national trends in
graduate enrollment, from Tech’s own undergraduate enrollment statistics and
from the long-term growth plans in the ECE department.
1.3.1 National PhD Trends
Nationally, of those universities
offering standalone doctoral programs in CpE, 5.2% of all engineering doctoral
degrees are in CpE [3].
1.3.2
Michigan
Michigan Tech undergraduate enrollment
statistics show similar trends at the baccalaureate level [4], as seen
nationwide at the PhD level. Figure 1 shows the contribution of the CpE
bachelor’s degree program to all first-time freshmen enrolled in the College of
Engineering. This figure shows that,
following the startup transient in 2001, the contribution of the CpE program
has oscillated about the 7% mark.
Figure 2 illustrates the impact of the
CpE BS program and other computing majors on each other, in terms of Freshman
Enrollment. In addition to CpE, CS, and EE, the chart includes Computer System
Science (CSS) and Software Engineering (SE) majors in the CS department. It
shows a startup transient in which CpE enrollment increased at the expense of
EE and CS enrollment. This was primarily due to “defections” from the other
majors into the new program. Following the startup transient, CpE enrollment
stabilized at about 25% of all computing majors.
1.3.3 Enrollment Extrapolation
ECE faculty members are currently
advising 10 PhD candidates in computer engineering topics under the auspices of
the EE PhD program. These students comprise 4.8% of engineering PhD students
and 15.4% of computing PhD students. This is about 93% of the national PhD
trend, about 59% of the Michigan Tech engineering majors trend, and 70% of the
Michigan Tech computing majors trend, based on undergraduate enrollment.
Initially, these current 10 CpE students will switch from the EE to the CpE PhD
program. Based on national and local
trends, it can be expected that initiation of a standalone PhD program in CpE
will attract up to eight additional students, reaching 18 PhD students, given the
current number of faculty.
1.3.4 Long-Term Strategic Planning
The proposed PhD program in Computer
Engineering is an integral part of the long-term strategic plan for the
Department of Electrical and Computer Engineering. The department is in the
first year of a three-year growth plan, developed as part of an agreement
between the university and the new department chair, Professor Daniel R.
Fuhrmann. Under this plan the department will grow from 21 to 26 tenured/tenure
track faculty over the next three years. These numbers do not include two
faculty members with tenure in the department who now hold administrative
positions in the College of Engineering.
As part of this plan we anticipate a net
growth of two faculty members in computer engineering. With one anticipated
retirement we will be searching for three CpE faculty members in all. This
growth plan is closely tied to Tech’s Strategic Faculty Hiring Initiative
(SFHI) in Computational Discovery and Innovation, which includes the Dave House
Professorship in Computer Engineering and potentially one or more other
positions in the department.
Concurrent with the growth in faculty
size is the planned growth in PhD programs. The department’s goal is to expand
graduate research activity to the point where we will have 12 PhD graduates per
year (EE and CpE combined) or roughly 0.5 PhD graduates per research-active faculty
member per year. Following our undergraduate trends, we anticipate that
one-third of these, or four PhD graduates per year, will come from the proposed
Computer Engineering program. We envision three graduates having advisors from
the core computer engineering faculty group with one other student having an
advisor from other areas of the department or even outside the department. With
a nominal 6:1 ratio for PhD enrolled students to graduates per year, the
department plans on 24 enrolled students in the CpE program. (Please see
section 3, Resource Requirements, for further discussion of the needed
resources.)
Note that while we see modest growth in
faculty to support the proposed program, we also recognize that the proposed
program is critical to our ability to attract and retain top-flight faculty in
computer engineering. Having a viable program in Computer Engineering is a
concrete indication of Tech’s commitment to this field and it will undoubtedly
play a role in the department’s hiring as well as the university’s SFHI in
Computational Discovery and Innovation.
2
Implementation
2.1
Planned Implementation
Date
This program is intended to begin as soon
as it is approved. At that time, currently enrolled students will be invited to
change degree programs to CpE, subject to eligibility. New students will be
accepted at the beginning of the first complete term following approval of this
program.
2.2
Scheduling Plans
This program will be implemented through
normally scheduled daytime classes.
2.3
Faculty
All
faculty in the Department of Electrical and Computer Engineering are
responsible for the success of all curricular programs in the department. Those
expected to be most involved in the proposed graduate degrees in computer
engineering include:
- Associate Professor Roger Kieckhafer specializes in fault-tolerant computing
systems, including safety critical, high assurance and high integrity
systems and the theoretical foundations of fault tolerance. He also
does reliability modeling and formal design methods and does interdisciplinary
research with the physics department.
- Associate Professor Tricia Chigan, an NSF Career
award winner, studies computer/communication networks and network
security. Her work includes
vehicular ad hoc networks, wireless ad hoc and sensor networks, wireless
network security, adaptive protocol design for cognitive radio networks,
dependable computing and communication systems, and network resource
allocation and management.
- Associate Professor Jindong
Tan’s research focuses on robotic sensor networks and body area sensor
networks; his research in robotic
sensor networks investigates the coordination of mobile robots and sensor
networks for coordinate sensing and communication. His research in body
sensor networks focuses on ultra lower design for a hybrid of
wearable, ingestible and implantable wireless miniature sensors, which
collectively monitor the medical condition of a patient and provide
physicians with immediate feedback. He collaborates with the
departments of Biomedical Engineering, Cognitive and Learning Sciences,
Computer Science, Civil and Environmental Engineering.
- Associate Professor Ashok
Goel models interconnects, currently for nanoscale integrated circuits. He
is the author of a highly acclaimed graduate text on VLSI interconnects.
- Assistant Professor Shiyan
Hu studies computer-aided design of
VLSI circuits and combinatorial optimizations. He focuses on nanoscale
circuit design challenges, such as interconnect optimization and design
for manufacturability.
- Assistant Professor Bo
Chen, whose primary appointment is in ME-EM, works on distributed sensing and
actuation systems. Current research of the Laboratory of Intelligent
Mechatronic and Embedded Systems concerns developing an autonomous network
framework for distributed sensing and actuation systems.
Curricula Vitae of the CpE faculty are
at: http://www.ece.mtu.edu/faculty/rmkieckh/Grad/
, http://www.ece.mtu.edu/faculty/goel/resume.htm, http://www.me.mtu.edu/meem/facultybio/b_chen.html.
In addition all ECE faculty of
professorial rank may be involved in the program. In addition to those faculty
listed above these faculty are Professors Dan Fuhrmann, Warren Perger, Mike
Roggemann, Martha Sloan, and Dennis Wiitanen; Associate Professors Paul
Bergstrom, Jeff Burl, Anand Kulkarni, John Lukowski, Bruce Mork, Gerry Tian and
Reza Zekavat; and Assistant Professors Chris Middlebrook, Wayne Weaver and Zack
Zhao.
Professorial rank faculty in Computer
Science who may be involved in the program are Professors Steve Carr and Linda
Ott; Associate Professors Jean Mayo, Phillip Merkey, Nilufer Onder, Soner
Onder, Dave Poplawski, Steve Seidel, C K Shene, and Chuck Wallace; and Assistant
Professors Ali Ebnenasir, Robert Pastel and Zhenlin Wang.
2.4
Curriculum Design
All requirements for the number of credits
required, credit distributions, theses, reports, examinations, acceptable
grades, time-to-degree, and other degree requirements are identical to existing graduate school requirements policies and
procedures [5], except as modified or enhanced by this proposal.
2.4.1 Admission Requirements
All applicants for full admission must have
completed courses in the set of prerequisite topics specified in Table 1. Applicants
who have not completed all of the prerequisites may receive “provisional”
admission and complete the missing topics at MTU. Those topics with an “MTU
Equivalent” course at less than the 3000 level may not be taken for PhD degree
credit, while those at or above the 3000 level may count for degree credit
within the constraints of all other applicable course distribution
requirements.
|
Table 1:
Prerequisite Courses |
|
|
Prerequisite
Topic |
MTU
Equivalent |
|
Linear
Algebra |
MA-2321 |
|
Differential
Equations |
MA-3521 |
|
Probability
and Statistics |
MA-3710
or 3720 |
|
Discrete
Math or Structures |
CS-2311 |
|
Data
Structures |
CS-2321 |
|
Computer
Organization |
CS-3421 |
|
Digital
Logic |
EE-2171
or 2173 |
|
Electronics |
EE-3130 |
|
Microcontroller
Interfacing |
EE-3170
or 3173 |
Given the prerequisite topics listed, students
with a Bachelor’s of Master’s degree in Computer Engineering from an accredited
college or university will generally be eligible for full admission to this
program. Those with a degree in Computer Science, Electrical Engineering, or a
closely related field, will usually be eligible for provisional admission. Applicants
with degrees from other disciplines may be considered for provisional admission
to the program on a case by case basis. The ECE department has an established
program of admitting graduate students with extenuating circumstances at the
discretion of the graduate program director under the provision that they
perform well their first year; their performance is tracked by the graduate
program director. Provisional students are not awarded GRAs or GTAs so that
they may concentrate on their studies.
2.4.2 Required Courses
All students must participate in EE 5970, Computer Engineering Seminar, 1 cr., Fall,
Spring.
In addition, all PhD students must complete
at least three Research Tools courses shown below within their first four semesters
in residence (not including summers). A student may be exempted from any of
these courses only if that student enrolls in a more demanding course in the
same topic within the four-semester deadline. The identity and total number of
courses required to be taken must be approved by the student’s advisor.
Table 2: Required Courses
|
EE 5970, Computer Engineering Seminar, 1 cr. |
|
|
three of the following four
courses |
|
|
EE 6710 – Research Tools for Computer Engineers—Experimental Statistics. 1 cr. |
|
|
EE 6711 – Research Tools for Computer Engineers—Stochastic Processes, Modeling and Simulation, 1 cr. |
|
|
EE 6712 – Research Tools for Computer Engineers—Formal Methods, 1 cr. |
|
|
EE 6713 – Research Tools for Computer Engineers—Hardware Implementation, 1 cr. |
2.4.3 Elective Course Distributions
In addition to graduate school
requirements for PhD course distributions [5] http://www.gradschool.mtu.edu/catalog/phd.html,
the following breadth criteria are
required for all CpE PhD degrees:
1.
At least 13
credits in the ECE department from the list of currently existing courses below,
including the required seminar
EE
4252 – Digital Signal Processing and Its Applications, 4 cr., Fall
EE
4255 – Wireless Communications, 3 cr.,
Spring
EE
4257 – Digital Image Processing, 3 cr., Spring
EE
4271 – VLSI Design, 4 cr., Fall, Summer
EE
4272 – Computer Networks, 3 cr., Fall
EE5522
– Digital Image Processing, 3 cr., Spring
EE
5525 – Wireless Communications, 3 cr., Spring
EE
5527 – Digital Communications, 3 cr., Fall
EE
5535 – Wireless Communications II, 3 cr., Spring
EE
5711 – Mathematical Techniques for Computer Networks, 3 cr., Fall
EE
5722 – Computer Networks, 3 cr., Fall
EE
5723 – Computer and Network Security, 3 cr., Spring
EE
5725 – Mobile Robotics & Mobile Robot Systems, 3 cr., On Demand
EE
5726 – Embedded Sensor Networks, 3 cr., On Demand
EE
5731 – Real-Time and Embedded Systems, 4 cr., On Demand
EE
5732 – Real-Time System Design, 4 cr., Fall, Spring
EE
5751 – Verilog HDL Design, 3 cr., Spring
EE
5752 – Digital Storage Technologies, 3 cr., Fall, even-numbered years
EE
5755 – Fault-Tolerant Systems, 3 cr., On Demand
EE
5772 – Parallel Computer Organizations, 3 cr., Fall, odd-numbered years
2.
At least 9
credits in the CS department, 4000-level or above.
CS 4090 – Special Topics in Computer Science, 3 cr., On Demand
CS 4121 – Programming Languages, 3 cr., Fall, Spring
CS 4131 – Compiler Construction, 4 cr., Spring
CS 4311 – Introduction to Computation Theory, 3 cr., Fall
CS 4321 – Introduction to Algorithms, 3 cr., Fall, Spring
CS 4331 – Introduction to Parallel Programming, 3 cr., On Demand
CS 4411 – Introduction to Operating Systems, 3 cr., Fall, Spring
CS 4421 – Database Systems, 3 cr., Spring
CS 4431 – Computer Architecture, 4 cr., Fall, Spring
CS 4481 – Computer and Network Performance Analysis, 4 cr., Spring
CS 4611 – Computer Graphics: Foundations
of Computer Graphics, 3 cr., Fall
CS 4710 – Model-Driven Software
Development, 3 cr., On Demand
CS 4711 – Software Processes and Management,
3 cr., Spring
CS 4712 – Software Quality Assurance, 3
cr., On Demand
CS 4760—Human-Computer Interactions, 3
cr., Spring
CS 4811 – Artificial Intelligence, 3
cr., Spring
CS 5090 – Special Topics in Computer
Science, 3 cr., On Demand
CS
5131 – Compiler Optimization, 3 cr., Fall, Spring
CS
5311 – Theory of Computation, 3 cr., On
Demand
CS
5321 – Advanced Algorithms, 3 cr., Fall
CS 5331 – Parallel Algorithms, 3 cr., Spring
CS
5411 – Advanced Operating Systems, 3 cr., Spring
CS
5431 – Advanced Computer Architecture, 3 cr., Spring
CS
5441 – Distributed Systems, 3 cr., Fall, Spring
CS
5461 – Mobile Networks, 3 cr., Fall, odd-numbered years
CS
5611 – Computer Graphics: Advanced Rendering and Animation, 3 cr., Spring
CS
5711 – Advanced Software Engineering, 3 cr., Fall
CS5760
– Human-Computer Interactions and Usability Testing, 3 cr., Spring
CS
5811 – Advanced Artificial Intelligence
CS
6090 –Special Topics in Computer Science, variable up to 3 cr., On Demand
CS
5461 – Advanced Computer Networks, 3 cr., Fall even-numbered years
2.4.4 Comprehensive Examinations
The comprehensive examination shall be
conducted in accordance with existing ECE Department procedures [6] for the EE
PhD program , with the subject matter customized for Computer Engineering
topics.
2.4.5 Program Completion Verification
Completion of all requirements shall be
certified by the student’s advisor using a degree audit form. The completed
form shall be forwarded to the ECE graduate program committee.
3
Resource
Requirements
3.1
Courses
These 1-credit research tools courses
will be introduced when the Ph.D. program begins.
EE 6710 – Research Tools for Computer Engineers—Experimental
Statistics
EE 6711 – Research Tools for Computer
Engineers—Stochastic Processes, Modeling and Simulation
EE 6712 – Research Tools for Computer
Engineers—Formal Methods
EE 6713 – Research Tools for Computer
Engineers—Hardware Implementation
3.2
Library and Other
Learning Resources
No new Library resources are required.
The library already subscribes to online versions of the leading journals in
this field.
3.3
Computing Access
Fee
The computing access fee will be the
same as for EE graduate degrees.
3.4
Available and Needed
Equipment
The Department
of Electrical and Computer Engineering has a full range of research
facilities including several Sun workstations and PCs running both Linux and MS
Windows. Each faculty member and graduate student has at least one PC, workstation
and/or laptop in his/her office and/or lab. Other PCs and servers are available
to students through several user laboratories throughout the building and
across campus. Faculty and students also have access to a high-speed Linux
cluster and a Beowulf cluster for research purposes. The Michigan Tech campus
is completely networked, allowing wired and wireless access to all services
from anywhere on campus, and to secure remote access via SSH, FTP and other
protocols. A wide range of research-relevant application software is also
available to both faculty and students.
No additional equipment is required for
this program.
3.5
Space
No additional space is required for this
program.
3.6
Accreditation
requirements
Not Applicable
3.7
Program Costs, Years
1, 2, and 3
No additional costs will be imposed, as
all faculty, equipment, and facilities are already in place for the initiation
of this plan.
3.8
Long-Range
Resource Needs
As
described in Section 1.3.4 above, the development of this PhD program in
Computer Engineering is an integral part of the long-term growth plans for the
Department of Electrical and Computer Engineering, as well as the Strategic
Faculty Hiring Initiative in Computational Discovery and Innovation. The university has already committed to the support
of hiring of three faculty members in
the computer engineering over the next three years. However, no resources beyond the typical
start-up packages for new faculty members, for which there is already a
university commitment, will be required.
As
part of its planned growth, the ECE department does intend to request
additional Graduate Assistant (GA) lines to remain at the current level of one
GA per faculty member. No commitment
from the Michigan Tech Graduate School has been given at this time regarding GA
lines.
References
1. Criteria for
Accrediting Engineering
Programs, Accreditation Board for
Engineering and Technology, Approved Nov 3, 2007, available at http://abet.org/Linked%20Documents-UPDATE
/Criteria%20and%20PP/E001%2008-09%20EAC%20Criteria%2012-04-07.pdf
2.
2006 Accreditation Statistics, Accreditation Board for Engineering and Technology,
available at http://abet.org/Linked%20Documents-UPDATE/Stats/06-AR%20Stats.pdf
.
3.
“Engineering Data
Management System Login”, American Society of Engineering Education, Apr 2008, http://www.asee.org/datamining/ .
4.
“Enrollment
Statistics”, Office of Student Records and Registration, Michigan Technological
University, Apr 2008, http://www.admin.mtu.edu/em/services/erlstat/
.
5.
“Doctor of
Philosophy”, Graduate School, Michigan Technological University, Apr 2008, http://www.gradschool.mtu.edu/catalog/phd.html
.
6.
“PhD
Comprehensive Examination”, Department of Electrical and Computer Engineering,
Michigan Technological University, Apr 2008, http://www.ece.mtu.edu/pages/graduate/PhD_Exam_Guidelines/index.htm
.
Adopted by Senate: 11 February 2009
Approved by Administration: 17 February 2009
Approved by BOC: 01 May 2009