Multimedia Communications

Directions and Innovations

Jerry D. Gibson

• Publisher: Academic Press
• Year: 2000
• Print ISBN:
  9780122821608
• eBook ISBN:
  9780080512242
• Topic: Technology & Engineering
• Subtopic: Computer Science General
• Index: Technology & Engineering

The rapid advances and industry demands for networked delivery of information and pictures through computer networks and cable television has created a need for new techniques and standards for the packaging and delivery of digital information. Multimedia Communications presents the latest information from industry and academic experts on all standards, methods and protocols. Internet protocols for wireless communications, transcoding of Internet multimedia for universal access, ATM and ISDN chapters, videoconferencing standards, speech and audio coding standards, multi-casting and image compression techniques are included. - Latest Internet protocols for wireless communications - Transcoding of Internet multimedia for universal access - ATM and ISDN chapters - Videoconferencing standards - Speech and audio coding standards - Multi-casting - Latest image compression techniques

CHAPTER 1

Multimedia Communications: Source Representations, Networks, and Applications

JERRY D. GIBSON

1.1 INTRODUCTION

Universal access to multimedia information is now the principal motivation behind the design of next-generation computer and communications networks. Furthermore, products are being developed to extend the capabilities in all existing network connections to support multimedia traffic. This is a profound paradigm shift from the original analog-voice telephony network developed by the Bell System and from the packet-switched, data-only origins of the Internet. The rapid evolution of these networks has come about because of new technological advances, heightened public expectations, and lucrative entrepreneurial opportunities.

In this chapter and in this book as a whole, we are interested in multimedia communications; that is, we are interested in the transmission of multimedia information over networks. By multimedia, we mean data, voice, graphics, still images, audio, and video, and we require that the networks support the transmission of multiple media, often at the same time. Two observations can be made at the outset. The media to be transmitted, often called sources, are represented in digital form, and the networks used to transmit the digital source representations may be classified as digital communications networks, even though analog modulation is often used for free-space propagation or for multiplexing advantages. In addition to the media sources and the networks, we will find that the user terminals, such as computers, telephones, and personal digital assistants (PDAs), also have a large impact on multimedia communications and what is actually achievable.

The development here breaks the multimedia communications problem down into the components shown in Figure 1.1. Components shown there are the Source, the Source Terminal, the Access Network, the Backbone Network, the Delivery Network, and the Destination Terminal. This categorization allows us to consider two-way, peer-to-peer communications connections, such as videoconferencing or telephony, as well as asymmetric communications situations, such as broadcasting or video streaming. In Figure 1.1, the Source consists of any one or more of the multimedia sources, and the job of the Source Terminal is to compress the Source such that the bit rate delivered to the network connection between the Source Terminal and the Destination Terminal is at least approximately appropriate. Other factors may be considered by the Source Terminal as well. For example, the Source Terminal may be a battery-power-limited device or may be aware that the Destination Terminal is limited in signal processing power or display capability. Further, the Source Terminal may packetize the data in a special way to guard against packet loss and aid error concealment at the Destination Terminal. All such factors impinge on the design of the Source Terminal. The Access Network may be reasonably modeled by a single line connection, such as a 28.8 Kbit/s modem, a 56 Kbit/s modem, a 1.5 Mbit/s Asymmetric Digital Subscriber Line (ADSL) line, and so on, or it may actually be a network that has shared capacity, and hence have packet loss and delay characteristics in addition to certain rate constraints. The Backbone Network may consist of a physical circuit-switched connection, a dedicated virtual path through a packet-switched network, or a standard best-effort Transmission Control Protocol/Internet Protocol (TCP/IP) connection, among other possibilities. Thus, this network has characteristics such as bandwidth, latency, jitter, and packet loss, and may or may not have the possibility of Quality of Service (QoS) guarantees. The Delivery Network may have the same general set of characteristics as the Access Network, or one may envision that in a one-to-many transmission that the Delivery Network might be a corporate intranet. Finally, the Destination Terminal may have varying power, mobility, display, or audio capabilities.

FIGURE 1.1 Components of a multimedia communications network.

The source compression methods and the network protocols of interest are greatly determined by international standards, and how these standards can be adapted to produce the needed connectivity is a challenge. The terminals are specified less by standards and more by what users have available now and are likely to have available in the near future. The goal is clear, however—ubiquitous delivery of multimedia content via seamless network connectivity.

We will first present discussions of the various components in Figure 1.1, and then we elaborate by developing common examples of multimedia communications and highlight the challenges and state-of-the-art. We begin our discussions with the Networks and Network Services.

1.2 NETWORKS AND NETWORK SERVICES

We focus in this section on everything between the Source Terminal and the Destination Terminal in Figure 1.1. Two critical characteristics of networks are transmission rate and transmission reliability. The desire to communicate using multimedia information affects both of these parameters profoundly. Transmission rate must be pushed as high as possible, and in the process, transmission reliability may suffer. This becomes even more true as we move toward the full integration of high-speed wireless networks and user mobility. A characterization of networks and network services according to rate is shown in Table 1.1. These networks and services not only show a wide variation in available transmission rates, but also the underlying physical transmission media vary dramatically, as do the network protocols. The additional considerations of wireless local area networks (LANs), cellular data, and mobility add a new dimension to network reliability, through the physical layer channel reliability, that makes the problem even more challenging.

Table 1.1

Networks and Network Services

Abbreviations: CATV, cable television; FDDI, Fiber Distributed Data Interface; OC-N/STS-N, optical cable-number of times the single link bandwidth/synchronous transport protocol-number of times the single link bandwidth; VDSL, very high rate digital subscriber line.

The Access and Delivery Networks are often characterized as the “last-mile” network connections and are often one of the first five entries in Table 1.1. Certainly, most people today connect to the Internet through the plain old telephone system (POTS) using a modem that operates at 28.8 Kbit/s up to 56 Kbit/s. While relatively low speed by today’s standards and for the needs of multimedia, these connections are reliable for data transmission. For transporting compressed multimedia, however, these lower speeds can be extremely limiting and performance limitations are exhibited through slow download times for images, lower frame rates for video, and perhaps noticeable errors in packet voice and packet video. Of course, as we move to the higher network speeds shown in the table, users experience some of the same difficulties if the rates of the compressed multimedia are increased proportionately or the number of users sharing a transport connection is increased. For example, when users move from POTS to Integrated Services Digital Network (ISDN) to ADSL, they often increase the rate of their multimedia transmissions and thus continue to experience some packet losses even though they have moved to a higher rate connection. Further, the higher speed connectivity in the Access Networks increases the pressure on the Backbone Networks or servers being accessed to keep up. Thus, even though a user has increased Access Network bandwidth, packet losses and delay may now come from the Backbone Network performance. Additionally, although POTS, ISDN, and ADSL connections are usually not shared, CATV services are targeted to support multiple users. Therefore, even though there is 20 Mbits/s or more available, a few relatively high rate users may cause some congestion.

Many people are seeking to upgrade their individual Access Network transmission rate, and higher speed modems, ISDN, Digital Subscriber Line (xDSL), and cable modems are all being made available in many areas. As users obtain higher Access Network rates, possible bottlenecks move to the Backbone Networks. This potential bottleneck can be viewed on a couple of levels. First, it is not unusual today for users to experience delays and congestion due to a lower rate Delivery Network or server. If delays are experienced when accessing a remote web site, the user does not know whether the difficulty is with the Access Network speed, the Backbone Network, the Delivery Network speed (in either direction), or the server being accessed (the server would be the Destination Terminal in Figure 1.1). Of course, commercial servers for web sites have a financial motivation for maintaining adequate network rates and server speeds.

Notice that there are substantial differences in the protocols used for several of the network services mentioned in Table 1.1; generally, these protocols were developed around the concept of transmitting non-time-critical data as opposed to time-sensitive multimedia traffic. Fortunately, the protocols have been designed to interoperate, so that network interfaces do not pose a problem for data-only traffic. The concept of internetworking is not too overwhelming if one considers only a set of isolated networks interconnected for the use of (say) one company. But when one contemplates the global internetwork that we call the Internet (capital I), with all of its diverse networks and subnetworks, having relatively seamless internetworking is pretty amazing. The Internet Protocol (IP) achieves this by providing connectionless, best-effort delivery of datagrams across the networks. Additional capabilities and functionality can be provided by transport layer protocols on top of IP. Transport layer protocols may provide guaranteed message delivery and/or correctly ordered message delivery, among other things.

In the Internet, the most commonly used transport layer protocol is the Transmission Control Protocol (TCP). TCP provides reliable, connection-oriented service and is thus well suited for data traffic as originally envisioned for the Internet [via Advanced Research Projects Agency Network (ARPANET)]. Unfortunately, one of the ways reliable delivery is achieved is by retransmission of lost packets. Since this incurs delay, TCP can be problematical for the timely delivery of delay-sensitive multimedia traffic. Therefore, for multimedia applications, many users often employ another transport layer protocol called User Datagram Protocol (UDP). Unlike TCP, UDP simply offers connectionless, best-effort service over the Internet, thus avoiding the delays associated with retransmission, but not guaranteeing anything about whether data will be reliably delivered.

When considering packet-switched networks like the Internet, we usually have in mind the situation where the Source Terminal wants to send packets to a single Destination Terminal. This is called unicast. There are situations, however, where the Source Terminal needs to send a message to all terminals on a network, and this is called broadcast. Since broadcasting sends one copy of the message for each end node or Destination Terminal, this type of transmission may flood a network and cause conge...

Table of contents

• Cover image
• Title page
• Table of Contents
• ACADEMIC PRESS SERIES IN COMMUNICATIONS, NETWORKING, AND MULTIMEDIA
• Copyright
• Dedication
• Preface
• List of Contributors
• Chapter 1: Multimedia Communications: Source Representations, Networks, and Applications
• Chapter 2: Future Telecommunication Networks: Traffic and Technologies
• Chapter 3: Speech Coding Standards
• Chapter 4: Audio Coding Standards
• Chapter 5: Still Image Compression Standards
• Chapter 6: Multimedia Conferencing Standards
• Chapter 7: MPEG-1 and -2 Compression
• Chapter 8: MPEG-4 and MPEG-7
• Chapter 9: ATM Network Technology
• Chapter 10: ISDN
• Chapter 11: Video-on-Demand Broadcasting Protocols
• Chapter 12: Internet Telephony Technology and Standards Overview
• Chapter 13: Wideband Wireless Packet Data Access
• Chapter 14: Internet Protocols Over Wireless Networks
• Chapter 15: Transcoding of the Internet’s Multimedia Content for Universal Access
• Chapter 16: Multicasting: Issues and Networking Support
• INDEX