Electrical Engineering Undergraduate Curriculum

 

 

Robert Dutton, Andrea Goldsmith, Nick McKeown, Eric Roberts,

Olav Solgaard, Dwight Nishimura, Ross Venook, Simon Wong, Mark Horowitz

 

1.     Introduction

The outline of the current undergraduate curriculum was put in place over 15 years ago, and while it has evolved, the basic structure and emphasis has remained unchanged.  Given the changing nature of both the field of Electrical Engineering and the background of our students, the department created a committee to review the curriculum.  This document is the report of that committee.  It starts by reviewing the state of the department and the current curriculum, and looks at the forces driving both the department and its students. Section 3 provides an overview of our revised curriculum and explains how this new plan adapts to both the changing background of our students, and the changing nature of EE . We propose to revise all of the classes that form the core of the curriculum, and Section 4 describes these new classes in more detail.  This section is followed by a description of depth sequences.  Section 6 describes an aggressive, but feasible, transition plan for moving to the new program, and Section 7 describes issues that we feel still need to be addressed in the future. 

 

While this report makes some very specific course proposals, we understand that the strength of a university like Stanford is enabling each faculty member to create the best class that they can.  Thus our course outlines are a method of being more concrete about what topics we think are core to an area, rather than rigidly defining the class.  Finally while we have spoken with many other faculty members to construct this document, our task is clearly not complete Ð we welcome further comments.  It is only through this feedback that we will create a successful EE curriculum for the 21st century.

 

2.     Background

As described in the most recent strategic plan of the Electrical Engineering Department, the field of Electrical Engineering has continued to evolved and expand, further blurring its boundaries with other disciplines. In particular the overlap with CS continues to grow, and joint undergraduate programs at the border with CS are essential. The ability to create joint programs  with other departments is strongly desired, with areas like bioelectronics worthy of serious consideration in the near future (although it was not considered in this report).

 

In addition to becoming broader, electrical engineeringElectrical Engineering has been strongly influenced by the rapid growth in information processing.  Information technology has both served both as a dominant consumer electronic technology, and provided the tools that drive further innovations.  As a consequence, the complexity of the systems that our students deal with has grown exponentially.  Our curriculum must provide them with not only the insights to understand the underlying technologies and theories associated with each level of complexity, but also the knowledge and skills to choose the appropriate abstraction level for each component appropriate for the task at hand, making the complexity work for them   rather than against them.

 

The rapid growth of information technology has also changed the background, training, and interests of our students. Gone are the days when prospective electrical engineeringElectrical Engineering students built or disassembled electronic systems (with radio / audio amplifiers being the most common) before they entered universities.  TodayÕs students have more exposure and background in software than hardware.  They have direct experience manipulating ÒcodesÓ but not ÒdevicesÓ, feel more at home in the virtual world of the computer, rather than the physical world.  Students are also used to dealing in a world with abundant information, and many distractions, and they feel more comfortable in situations where the application for the information being taught is clear.  Our current curriculum lays out the fundamentals first before getting to applications and is a ÔÒpoor impedance matchÕ matchÓ to our students.

 

In addition to delaying gratification, the sequence structure of the current curriculum causes additional problems for our students.  The core is too long, and too linear, making it difficult for students to create a schedule that allows them to take many classes in their depth area.  Some depth areas are hard to complete if you decide on an EE major latebelatedly. Some students take the EE111-113 sequence concurrently with EE101-103, to avoid some of these problems, which causes a different set of challenges for them. In addition this structure does not encourage a student to ÒsampleÓ different areas since a student needs to take many classes before reaching the essence or excitement of the area.  If we want to foster work at the border of different areas, we need to create classes that encourage sampling and build excitement in multiple areasclasses that build excitement in the first class of a series.

 

In summary,  we need to change our undergraduate curriculum to

á      motivate students to ÒsampleÓ different areas,

á      emphasize the integration of knowledgehow fundamental principles cut across different core areas,

á      include motivating examples for all the material in the core,

á      take advantage of the studentsÕ familiarity with ÒvirtualÓ environments,

á      arouse the studentsÕ interest and curiosity in Òhardware,Ó

á      blur the boundary between ÒsoftwareÓ and Òhardware,Ó

á      broaden the studentsÕ appreciation of system issues, and

á      familiarize students with different levels of system abstraction.

 

 

Unfortunately we need to implement these changes in a constrained environment. Stanford prides itself on being a liberal-arts university. Our undergraduates are not required to declare a major at the time of their admission, and have a number of distribution requirements during their first two years.  They are encouraged to explore and develop a variety of interests before choosing a major.  This both forces us to compete with other departments for the best students, and limits the amount of classes that we can include in our program.  We are faced with a small number of classes we can require all students to take. To make room for classes that help with abstractions and dealing with complex systems, some material needs to be dropped from the current core.   This is a difficult question, since all areas have strong proponents.

 

Our proposal keeps the core small, and uses it to introduce areas that are not covered in depth.  Our present curriculum was created when solid-state electronics was the key area in EE, and so the curriculum is centered on microelectronics devices and circuit design.  While this remains a key area, it no longer holds the dominant position it once did.  Thus we are reducing the number of courses in the solid-state electronics area, and broadening the remaining classes.  The next section outlines the changes that we propose.

 

3.     Key Features of the Proposed Curriculum

 

We decided to focus  our effort on defining the classes that would form the core of the undergraduate program Ð classes that every EE student must take Ð and classes that were are under the control of the EE faculty. While we all thought there were issues with the current Math and Physics requirements, we felt the current requirements were acceptable for the new program, and changing them was more an issue for the School of Engineering. We agreed that programming methodology (E70x Ð CS106x, or CS106B) should be kept as part of the program. 

 

To address many of the problems mentioned in the previous section, we set up the following goals for the development of the EE core requirements: