Issue No.04 - Winter (1995 vol.12)
Published by the IEEE Computer Society
Computer and communications companies have focused intensely on multimedia systems because they represent an expanding new market. One major semiconductor company estimates the total annual market for multimedia-related products will be 17 billion dollars by 2000. This fuels research in the design and test of such systems and drives changes in all aspects of system design. The demanding real-time, high-bandwidth requirements of video, audio, and interactive graphics affect application integration, display, computation, storage, and transmission.
Video is the most demanding and visually compelling component of multimedia. The advent of digital video as a ubiquitous medium compels us to understand its system development ramifications. This special issue examines design examples of multimedia systems and issues relating to transmitting and decoding video, audio, and graphical information. The four articles I've selected discuss key technical challenges and offer insight into research within industry and academia.
In the first article, I and coauthors Vasudev Bhaskaran and Devendra Kalra summarize how Hewlett-Packard Laboratories redesigned a processor for multimedia. From this example of where we are today, we go on to discuss what's on the horizon in this area. Innovations in conventional desktops—such as personal computers and workstations with modern RISC processors—will lead to next-generation consumer platforms known as digital set-top boxes.
We continue our design examples with a case study. University of Washington authors Jihong Kim and Yongmin Kim present the design and implementation of a highly integrated simulation environment based on the Multimedia Video Processor (MVP). Using tools they describe, designers built an actual multimedia system based on this processor. This case study is proof of the simulator's power to accurately model a highly integrated, real-world design.
Since transmitting video data usually involves compression and decoding, assessing image fidelity is important. That's why our third article describes a method for objectively assessing MPEG-2 coding fidelity. This work is the joint effort of Christian J. van den Branden Lambrecht and Murat Kunt of the Swiss Federal Institute of Technology, Lausanne; A1 Kovalick of Hewlett-Packard's Video Communications Division; and Vasudev Bhaskaran. They provide insightful examples of image fidelity problems and propose real-time sequences synthesized expressly for exposing coding artifacts in MPEG-based systems.
This special issue concludes with an article that describes how the existing telecommunications environment will transmit multimedia information. A television engineer at Tektronix, David K. Fibush underscores the importance of transport protocol standards to developers of multimedia systems. Because most data will conform to such emerging standards, it is important for system developers to understand them. Although the article primarily discusses digitally compressed audio and video data, it devotes attention to analog issues as well.
We currently have or are on the verge of having the technology to implement products such as digital televisions, set tops, and real-time, interactive multimedia computers. However, a major hurdle (and design focus) will be producing these products at a cost that the average consumer will pay. In addition, we must develop the infrastructure to support multimedia systems, which includes accepted standards, content providers, and a business model (that is, how do you get people to pay for it?) With developments such as the plethora of "media processors" that have been announced and interactive television, multimedia is here to stay and will engender even more design and test issues in the future.