ECE ECE594: Internet Computing and Digital TV
Winter Quarter 2000
Edward Chang

Enrollment Limit: 15
| Introduction | Lecture Outline | Projects | Prerequisites | Class Resources |

Introduction

ECE594 is an experimental course on Internet and Digital TV technologies. The course will cover the key technologies that enable the Internet and Digital TV and will also ask you to develop new technologies to experiment distributed lectures. This course will be co-taught by professor Edward Chang at UCSB and professor Pat Hanrahan at Stanford University via an Internet2 link. The iRoom at Stanford and the ECE conference room (room #2162) at UCSB will be connected with a high-speed (20 Mbps) Internet2 link. We plan to set up only very basic AV equipment at the start of the quarter. The students enrolled in the class will participate in projects that enhance the quality of the lectures. This is one of the first distributed classroom projects that is being experimented.

ECE594 is a project course with about nine lectures, including guest speakers from Intel, TIVO, WebTV, RealMedia (at Stanford), and NEC (at UCSB). The course will meet twice a week:
Monday 2pm to 3:50pm at Engineering-1 #2162 and
Wednesday 2pm to 3pm at Phelps #1437.
On Mondays, we will conduct joint-lectures with Stanford. On Wednesdays we will conduct local lectures and discuss course projects.

Vision

Linking two classrooms via an Internet connection is perhaps the easiest part of a distributed classroom project and everyone can do it. Making a distributed lecture immersive and interactive, however, requires much more than just having a connection. We plan to develop an array of key technologies to improve the quality of a distributed lecture.

Lecture Outline

Course Outline (HTML).

Projects

Some suggested projects are:
  1. Change Linux O.S. scheduler to support real-time periodic-task scheduling.
  2. Implement a set-top box that can play back multiple DTV streams simultaneously with minimum memory requirement and minimum delay to switch between the streams.
  3. Implement a video indexing scheme that can parse a live TV program and generate temporal/spatial/semantic indexes on the fly.
  4. Optimize an MPEG-2 decoder obtained from an open available source.
Project Groups
Project Proposal Template

Prerequisites

Strong programming skills and background in Operating Systems, Signal Processing or Database Systems.

Resources

    Stanford CS448a
    Ed's Thesis Presentation
    DTV Pointers(Courtesy of Milton Chen)
    Signal Processing Pointers(Courtesy of Robert Gray)