EECS 224: Digital Communications, Spring 2003

University of California at Berkeley
Dept of Electrical Engineering & Computer Sciences

[ Announcements | Administrative Info | Course Info | Handouts
Homework | Miscellany ]

Announcements

Rather than having a traditional written final exam on May 21, students will have to give an oral presentation about either a research topic of their choice or on a project implementing a communication system.

Special Announcement: Due to an NSF/FCC meeting in D.C., Prof. Sahai will not be able to hold office hours or lecture this Tuesday Feb 18th. Class has been postponed till Thursday Feb 20th. Update: That meeting was snowed out.

Slides of Dr. Gallager's talk, "THE IMPACT OF CLAUDE SHANNON ON MODERN DIGITAL COMMUNICATION"

Administrative Info

Instructor: Anant Sahai, Room 267 Cory Hall, sahai@eecs.berkeley.edu

Office hours: 2-4pm Tuesdays

Teaching Assistant: None

Course website: http://inst.EECS.Berkeley.EDU/~ee224/

Lectures: T-Th 12:30-2:00PM in 293 Cory Hall

Course Info

Description

The modern field of digital communication was pioneered by C. E. Shannon in 1948. Digital communication systems have now become the basic workhorses behind the information age. Examples include wireless and wireline telephone transmission systems, storage systems, telephone modems, cable modems, digital subscriber loop technology, etc. This course is an introduction to the fundamental principles underlying the design and analysis of digital communications systems.

Prerequisite:

The basic prerequisites for this course are familiarity with EE 120 (Signals and Systems) and EE 226A (Random Processes in Systems) or equivalent. In particular, familiarity with the following concepts are expected:

jointly Gaussian random vectors and Gaussian processes;
basic concepts in hypothesis testing and optimal detection;
signals viewed as elements of vector spaces.

Good references for this set of materials are my lecture notes and the reader I used last semester in 226A. Chapter two of the textbook and the handouts this term will also contain some review material. If you are not sure about your background, please come and talk with me.

Relationship with other courses:

There are several other communication-related courses offered in the Department:

EE 229: Information Theory — A comprehensive and in-depth treatment of the fundamental limits of communications, including both channel and source coding. The present course only touches upon information theory and will not cover source coding.
EE 290S: Coding Theory — The coding field is vast. This course will only cover the basic concepts. EE 290S will give a more in-depth and broader treatment. There are also sometimes courses in CS that touch upon coding.
EE 290S/224b: Wireless Communications — David Tse offered a course on wireless communication in Fall 2002 and expects to offer it again within the next two years. While I will spend a few weeks on wireless communication in 224, the 290S course on wireless will focus exclusively on wireless, emphasizing the modern developments in the field.
290S: Nontraditional Communication and Control — I was planning on offering an advanced special topics course this term on interactive communication in non-point-to-point settings and the relationship with control. This has been postponed till next term.

Requirements:

There will be regular problem sets (25% of course grade), a midterm (25%), and a final (50%).

Required text:

J. G. Proakis, Digital Communications, New York: McGraw-Hill, 4th ed., 2001.

Supplementary books

J. M. Wozencraft and I. M. Jacobs, Principles of Communication Engineering, Wiley, 1965. (Excellent book that popularizes the signal space approach to communication.)

E. A. Lee and D. G. Messerschmitt, Digital Communication, Kluwer Academic Publishers, 2nd ed., 1993.

Course Overview

In this course, we will start with an overview and then cover the basics of source coding and quantization. From there, we will move on to the basics of modulation and receivers emphasizing the signal space approach and start studying the probability of error. We will continue with the techniques used to communicate over ideal bandlimited Gaussian channels and use the complex discrete-time baseband representation.

With the basics of uncoded transmission in hand, we will take a minor digression to introduce the basic concepts from information theory like channel capacity and start our brief study of coding and coded modulation. With that done, we will continue with a look at what happens when the channel is not an ideal bandlimited channel and study equalization and how to combat inter-symbol-interference. Finally, we will close out the course with a look at the issues that arise in mobile wireless communication and how to think about time variation.

This course will emphasize modeling and how to think about problems in communication by breaking them down into a sequence of sub-problems. As such, examples will be used throughout the class.