AVI - Audio Video Interleaved. A Microsoft format for digital audio and video playback from Windows 3.1 Somewhat cross-platform, but mostly a Windows format. Has been replaced by the ASF format, but still used by some multimedia developers.

bandwidth - A measure of the amount of data that can transmit over a network (e.g., in bits per second). For example, a regular Ethernet line has a bandwidth of 10 Mbps, a wireless IEEE 802.11b 11 Mbps. Note that bandwidth is sometimes measured as the bandwidth in one direction, and sometimes as the total in both directions.

bit rate - number of data bits per second that can flow in a communications circuit (e.g., up to 56K for dial up modems, 384K for 3-ISDN videoconferencing).

CIF - A video format that supports both NTSC and PAL signals. CIF is part of the ITU H.261 videoconferencing standard. It specifies a data rate of 30 frames per second (fps), with each frame containing 288 lines and 352 pixels per line.

CODEC - Coder/decoder (a telecommunications term) or compressor/decompressor (a computer term). It is hardware or software that compresses and decompresses digital data/audio/video signals.

chrominance - The portion of a video signal that specifies what color each portion of the picture is to be. See also Luminance, S-Video and Composite Video.

Composite video - A method of carrying video information, which combines chrominance and luminance on a single wire, resulting in lower video quality than S-Video

decoder - A piece of hardware or software that is used to convert video or audio (typically) from the digital form used in transmission or storage into a form that can be viewed.

echo suppression - Echo suppression is a crucial portion of all video conferencing systems. If echo is not suppressed, the speaker hears his own audio coming back from the other end of the circuit, after a small time delay. Th fault always lies with the far end, although they do not perceive any problem.

full duplex - Sending data in both directions at the same time. Usually higher quality but requires more bandwidth. In video conferencing, full duplex will be much more natural and useable. Cheap speakerphones are half duplex, whereas more expensive ones are full duplex.

G.7xx - A family of ITU standards for audio compression.

gatekeeper - In the H.323 world, the gatekeeper provides several important functions. First, it controls access to the network, allowing or denying calls and controlling the bandwidth of a call. Second, it helps with address resolution, making possible email type names for end users, and converting those into the appropriate network addresses. They also handle call tracking and billing, call signaling, and the management of gateways. They also handle call tracking and billing, call signaling, and the management of gateways.

gateway - Gateways provide a link between the H.323 world and other video conferencing systems. A common example would be a gateway to a H.320 (ISDN) video conferencing system.

H.261 - ITU standard for video coding for videoconferencing. H.261 is a discrete cosine transform (DCT) based algorithm for video in the 64kb/s to 2mb/s range. All H.323 compliant video conferencing system are required to support this codec.

H.263 - ITU standard for video coding within videoconferencing. H.263 offers better compression than H.261, particularly in the low bitrate range used by modems.

H.320 - ITU standard for videoconferencing over ISDN and fractional T1 lines.

H.323 - ITU standard for videoconferencing over networks that do not guarantee bandwidth, such as the Internet. H.323 is the standard that this cookbook is recommending that most users in the education community should be using. For more detailed information on this and the other ITU standards see the bibliography of this document.

H.324 - ITU standard for video conferencing over standard phone lines.

half duplex - A telecommunications system where data can only flow in one direction at a time. Cheaper speakerphones are a good example of this, where only one person can talk at a time.

IETF - Internet Engineering Task Force. This is a group that develops and publishes new standards for use on the Internet.

IP - The Internet Protocol. IP is the basic language of the Internet. It was developed by the government for use in internetworking multiple computer networks together.

IP Multicast - A system for sending IP transmissions out only one time, but allowing for multiple users to receive it. This would reduce the bandwidth required for audio and video broadcasting over the Internet, but it is not widely used yet.

jitter - The change in latency with time. This is a network problem that is very important to video quality. Significant jitter destroys video.

latency - The length of time it takes a packet to move from source to destination; delay.

lossless compression - Refers to data compression techniques in which no data is lost. For most types of data, lossless compression techniques can reduce the space needed by only about half. Only certain types of data can tolerate lossy compression. Lossless compression technique when compressing data and programs.

lossy compression - Refers to data compression techniques in which some amount of data is lost. Lossy compression technologies attempt to eliminate redundant or unnecessary information. Most video compression technologies, such as MPEG, use a lossy technique.

luminance - The portion of a video signal that specifies how bright each portion of the picture is to be. See also Chrominance, S-Video and Composite Video.

MIDI - Musical Instrument Digital Interface is a standard for connecting electronic musical instruments and computers. MIDI files can be thought of as digital sheet music, where the computer acts as the musician playing back the file. MIDI files are much smaller than digital audio files, but the quality of playback will vary from computer to computer.

MPEG - MPEG (Moving Picture Experts Group) is a series of ISO standards for digital video and audio, designed for different uses and data rates.

Multipoint Conferencing Server (MCS) (also MCU) - A hardware or software H.323 device that allows multiple video conferencing (or audio or data) users to connect together. Without an MCS typically only point-to-point conferences can take place. Commonly supports voice activated switching, where whoever is talking is broadcast to all users. ITU-T standard H.231 describes the standard way of doing this. Many current systems only support H.320 (ISDN) but many vendors are working to upgrade their products to support H.323 (LAN, Internet) as well. In the H.320 space, this functionality is referred to as a multipoint control unit (MCU). Sometimes these terms are used interchangeably, although they refer to somewhat different implementations.

packet - A unit of information sent across a (packet-switched) network. A packet generally contains the destination address as well as the data to be sent.

QCIF - A standard related to CIF, QCIF (Quarter CIF), transfers one fourth the amount of data and is suitable for videoconferencing systems on slower connections or telephone lines.

streaming media - Sending one-way video or audio over a network as needed, instead of forcing the user to download the entire file before viewing it. Typically a few seconds of data is sent ahead and buffered in case of network transmission delays. (Although some data is buffered to the hard drive, it is written to temporary storage and is gone once viewing is complete.)

S-Video - A method of carrying video information on a cable that separates luminance and chrominance on separate wires, thereby providing higher video quality than composite video. See also Chrominance, Luminance and Composite Video.

T.120 - T.120 is an ITU-T standard (International Telecommunications Union) for document conferencing. Document conferencing allows two or more people to concurrently view and edit a document across a network. T.120 is the commonly used name to refer to a family of distinct standards, including chat andfile transfer. Many video conferencing companies were developing their own implementations of this until Microsoft released its free NetMeeting software. Now, many companies are using NetMeeting, often using its SDK to enhance it in some way.

Teleconferencing - Two or more people who are geographically distant having a meeting of some sort across a telecommunications link. Includes audio conferencing, video conferencing, and or data conferencing.

Terminal End Station - A terminal end station is the client endpoint that provides real-time, two-way communications. This is often shortened to just terminal.

Transcoding - Converting a data stream from one format to another, such as MPEG 1 to H.263, or an H.320 videoconferencing session to H.323.

unicast - Sending each user their own copy of a video (or other data) stream. As opposed to Multicast, where one copy is sent and whoever wants it listens to that copy. It is the most commonly used method for video conferencing and video on demand today. Multicast, which is much more efficient, is slowly gaining ground, but requires Internet Service Providers to support it.

video-on-demand - Being able to view any of a number of videos when you want to. Used on the internet and at hotels, cable systems, etc.

video server - A computer server that has been designed to store large amounts of video and stream it to users as required. Usually a video server has large amounts of high-speed disks and a large amount of network bandwidth to allow for many users to simultaneously view videos.

voice activated switching - Automatically switching the video feed to whoever is speaking in a multipoint videoconference. Usually a function of the MCU (multipoint conferencing unit)