US 5,774,674 AGrant

System for Negotiating at Least Two Sets of Video Capabilities Between Two Nodes to Perform Video Conferencing Between the Nodes According to the Selected Set

Issue Date:1998-06-30

•24 Claims

•57 Drawing Sheets

Abstract

Nodes of a computer-based teleconferencing network may be able to encode and decode video signals using different video compression/decompression algorithms with different parameters. In order to conduct a video teleconference between two (or more) such nodes, the nodes must first agree on what video algorithms to use. In the present invention, video capabilities are negotiated between nodes, where each node uses a table that identify one or more sets of video capabilities for that node. After video capabilities have been successfully negotiated, the video conferencing can proceed using the negotiated capabilities. Negotiated video capabilities may include algorithms, frame rate, frame resolution, bit rate, and bitstream format.

Metadata

Assignee

Intel Corporation

Inventors

Michael Gutmann
Katherine Cox
Reed Sloss

Application Information

Application Number:US 4792947

Filing Date:1995-06-07

Priority Date:1993-11-24

Art Unit:237

Classifications

IPC:

G06F 1300

Field of Search:

395370379348200.01;200.02;200.04;200.05;200.09;200.34;200.5;200.52;200.55;200.57;200.58 ;200.6;200.67;33062;260;26120216

Patent Drawings (57 sheets)

Description

This application is related to U.S. patent application Ser. No. 08/342,076 (filed Nov. 16, 1994, now abandoned), U.S. patent application Ser. No. 08/305,206 (filed Sep. 13, 1994, now U.S. Pat. No. 5,600,684), U.S. patent application Ser. No. 08/137,319 (filed Oct. 14, 1993, now U.S. Pat. No. 5,452,299), U.S. patent application Ser. No. 08/170,146 (filed Dec. 20, 1993, now U.S. Pat. No. 5,581,702), U.S. patent application Ser. No. 08/235,955 (filed Apr. 28, 1994, now U.S. Pat. No. 5,493,514), and U.S. patent application Ser. No. 08/133,612 (filed Oct. 12, 1993, now U.S. Pat. No. 5,410,698), which are all incorporated herein in their entireties by reference.

Background of the Invention

1. Field of the Invention

The present invention relates to audio/video conferencing, and, in particular, to systems for real-time audio, video, and data conferencing in windowed environments on personal computer systems.

2. Description of the Related Art

It is desirable to provide real-time audio, video, and data conferencing between personal computer (PC) systems operating in windowed environments such as those provided by versions of Microsoft.RTM. Windows.TM. operating system. There are difficulties, however, with providing real-time conferencing in non-real-time windowed environments. It is also desirable to provide conferencing between PC systems over two or more different transports.

It is accordingly an object of this invention to overcome the disadvantages and drawbacks of the known art and to provide real-time audio, video, and data conferencing between PC systems operating in non-real-time windowed environments over two or more different transports.

It is a particular object of the present invention to provide real-time audio, video, and data conferencing between PC systems operating under a Microsoft.RTM. Windows.TM. operating system over ISDN and LAN networks.

Further objects and advantages of this invention will become apparent from the detailed description of a preferred embodiment which follows.

Summary of the Invention

The present invention comprises a computer-implemented process and apparatus for video conferencing. According to a preferred embodiment, video capabilities are negotiated by a first node with a second node, wherein the first node bases its negotiating on a first-node table that identifies one or more sets of video capabilities for the first node. One of the sets of video capabilities for the first node is selected for video conferencing between the first and second nodes. Local encoded video signals are generated by the first node in accordance with the selected set of video capabilities. The local encoded video signals are transmitted from the first node to the second node. Remote encoded video signals are received by the first node from the second node. The remote encoded video signals are decoded by the first node in accordance with the selected set of video capabilities.

Description of the Drawings

Other objects, features, and advantages of the present invention will become more fully apparent from the following detailed description, the appended claims, and the accompanying drawings in which:

FIG. 1 is a block diagram representing real-time point-to-point audio, video, and data conferencing between two PC systems;

FIG. 2 is a block diagram of the hardware configuration of the conferencing system of each PC system of FIG. 1;

FIG. 3 is a block diagram of the hardware configuration of the video board of the conferencing system of FIG. 2;

FIG. 4 is a block diagram of the hardware configuration of the audio/comm (ISDN) board of the conferencing system of FIG. 2;

FIG. 5 is a block diagram of the software configuration of the conferencing system of each PC system of FIG. 1;

FIG. 6 is a block diagram of the hardware configuration of the audio/comm (ISDN) board of FIG. 4;

FIG. 7 is a block diagram of the conferencing interface layer between the conferencing applications of FIG. 5, on one side, and the comm, video, and audio managers of FIG. 5, on the other side;

FIG. 8 is a representation of the conferencing call finite state machine (FSM) for a conferencing session between a local conferencing system (i.e., caller) and a remote conferencing system (i.e., callee);

FIG. 9 is a representation of the conferencing stream FSM for each conferencing system participating in a conferencing session;

FIG. 10 is a representation of the video FSM for the local video stream and the remote video stream of a conferencing system during a conferencing session;

FIG. 11 is a block diagram of the software components of the video manager of the conferencing system of FIG. 5;

FIG. 12 is a representation of a sequence of N walking key frames;

FIG. 13 is a representation of the audio FSM for the local audio stream and the remote audio stream of a conferencing system during a conferencing session;

FIG. 14 is a block diagram of the architecture of the audio subsystem of the conferencing system of FIG. 5;

FIG. 15 is a block diagram of the interface between the audio task of FIG. 5 and the audio hardware of audio/comm (ISDN) board of FIG. 2;

FIG. 16 is a block diagram of the interface between the audio task and the comm task of FIG. 5;

FIG. 17 is a block diagram of the comm subsystem of the conferencing system of FIG. 5;

FIG. 18 is a block diagram of the comm subsystem architecture for two conferencing systems of FIG. 5 participating in a conferencing session over an ISDN connection;

FIG. 19 is a representation of the comm subsystem application FSM for a conferencing session between a local site and a remote site;

FIG. 20 is a representation of the comm subsystem connection FSM for a conferencing session between a local site and a remote site;

FIG. 21 is a representation of the comm subsystem control channel handshake FSM for a conferencing session between a local site and a remote site;

FIG. 22 is a representation of the comm subsystem channel establishment FSM for a conferencing session between a local site and a remote site;

FIG. 23 is a representation of the comm subsystem processing for a typical conferencing session between a caller and a callee;

FIG. 24 is a representation of the structure of a video packet as sent to or received from the comm subsystem of the conferencing system of FIG. 5;

FIG. 25 is a representation of the compressed video bitstream for the conferencing system of FIG. 5;,

FIG. 26 is a representation of a compressed audio packet for the conferencing system of FIG. 5;

FIG. 27 is a representation of the reliable transport comm packet structure;

FIG. 28 is a representation of the unreliable transport comm packet structure;

FIG. 29 are diagrams indicating typical TII-DLM connection setup and teardown sequences;

FIGS. 30 and 31 are diagrams of the architecture of the audio/comm (ISDN) board;

FIG. 32 is a diagram of the audio/comm (ISDN) board environment;

FIG. 33 is a flow diagram of the on-demand application invocation processing of the conferencing system of FIG. 5;

FIG. 34 is a flow diagram of an example of the processing implemented within the conferencing system of FIG. 5 to manage two conferencing applications in a single conferencing session with a remote conferencing system;

FIG. 35 represents the flow of bits between two remote high-resolution counters used to maintain clock values over a conferencing network;

FIG. 36 is a flow diagram of the processing of the conferencing system of FIG. 1 to control the flow of signals over reliable channels;

FIG. 37 is a flow diagram of the preemptive priority-based transmission processing implemented by the communications subsystem of the conferencing system of FIG. 1;

FIG. 38 is a state diagram for the complete rate negotiation processing;

FIG. 39 is a state diagram for the rate negotiation processing for a called node during a 64 KBPS upgrade;

FIG. 40 is a state diagram for the rate negotiation processing for a calling node during a 64 KBPS upgrade; and

FIG. 41 is a state diagram for the rate negotiation processing in loopback mode during a 64 KBPS upgrade;

FIG. 42 is a flow diagram of the processing by the conferencing system of FIGS. 5 and 17 during the automatic transport detection implemented at install time;

FIG. 43 is a block diagram showing the network connections made by the conferencing system of FIGS. 5 and 17 during the automatic transport detection implemented at run time;

FIG. 44 is a representation of the DLMLAN packet header format;

FIG. 45 is a representation of the MDM packet header format for LAN transmissions;

FIG. 46 is a representation of the connection messages for a typical conferencing session from the perspective of the MDMs on the local and remote nodes;

FIG. 47 is a flow diagram of the video negotiation processing between two conferencing systems of FIG. 1;

FIG. 48 is a flow diagram of the call-progress processing when the placement of a conference call is successful;

FIG. 49 is a representation of the interrupt-time processing for receiving data signals by the audio/video conferencing system of FIG. 5;

FIG. 50 is a representation of the interrupt-time processing for transmitting data signals by the audio/video conferencing system of FIG. 5;

FIG. 51 is a representation of the auto registration environment for video conferencing;

FIG. 52 is a representation of the architecture for auto registration and remote confidence testing for the new node of FIG. 51;

FIG. 53 is a flow diagram of the processing for the auto registration and remote confidence testing of the auto registration environment of FIG. 51;

FIG. 54 is a flow diagram of the processing implemented by the client (i.e., a new node) for the auto registration processing of FIG. 53;

FIG. 55 is a flow diagram of the processing implemented by a confidence test server for the auto registration processing of FIG. 53;

FIG. 56 is a representation of the auto registration file format; and

FIG. 57 are connection diagrams that show the interactions between a DLM and an MDM in connection and session establishment and tear-down.

Description of the Preferred Embodiment(s)

Point-To-Point Conferencing Network

Referring now to FIG. 1, there is shown a block diagram representing real-time point-to-point audio, video, and data conferencing between two PC systems, according to a preferred embodiment of the present invention. Each PC system has a conferencing system 100, a camera 102, a microphone 104, a monitor 106, and a speaker 108. The conferencing systems communicate via network 110, which may be either an integrated services digital network (ISDN), a local area network (LAN), or a wide area network (WAN). Each conferencing system 100 receives, digitizes, and compresses the analog video signals generated by camera 102 and the analog audio signals generated by microphone 104. The compressed digital video and audio signals are transmitted to the other conferencing system via network 110, where they are decompressed and converted for play on monitor 106 and speaker 108, respectively. In addition, each conferencing system 100 may generate and transmit data signals to the other conferencing system 100 for play on monitor 106. The video and data signals are displayed in different windows on monitor 106. Each conferencing system 100 may also display the locally generated video signals in a separate window.

Camera 102 may be any suitable camera for generating NSTC or PAL analog video signals. Microphone 104 may be any suitable microphone for generating analog audio signals. Monitor 106 may be any suitable monitor for displaying video and graphics images and is preferably a VGA monitor. Speaker 108 may be any suitable device for playing analog audio signals and is preferably a headset.

Conferencing System Hardware Configuration

Referring now to FIG. 2, there is shown a block diagram of the hardware configuration of each conferencing system 100 of FIG. 1. Each conferencing system 100 comprises host processor 202, video board 204, audio/comm (ISDN) board 206, LAN board 210, and ISA bus 208.

Referring now to FIG. 3, there is shown a block diagram of the hardware configuration of video board 204 of FIG. 2. Video board 204 comprises industry standard architecture (ISA) bus interface 310, video bus 312, pixel processor 302, video random access memory (VRAM) device 304, video capture module 306, and video analog-to-digital (A/D) converter 308.

Referring now to FIG. 4, there is shown a block diagram of the hardware configuration of audio/comm (ISDN) board 206 of FIG. 2. Audio/comm (ISDN) board 206 comprises ISDN interface 402, memory 404, digital signal processor (DSP) 406, and ISA bus interface 408, audio input/output (I/O) hardware 410.

LAN board 210 of FIG. 2 may be any conventional LAN card that supports standard driver interfaces and is preferably an Intel.RTM. EtherExpress.TM. 16C LAN Combo Card.

Conferencing System Software Configuration

Referring now to FIG. 5, there is shown a block diagram of the software configuration each conferencing system 100 of FIG. 1. Video microcode 530 resides and runs on pixel processor 302 of video board 204 of FIG. 3. Comm task 540 and audio task 538 reside and run on DSP 406 of audio/comm (ISDN) board 206 of FIG. 4. The one or more network stacks 560 reside and run partially on host processor 202 of FIG. 2 and partially on LAN board 210 of FIG. 2. All of the other software modules depicted in FIG. 5 reside and run on host processor 202.

Video, Audio, and Data Processing

Referring now to FIGS. 3, 4, and 5, audio/video conferencing application 502 running on host processor 202 provides the top-level local control of audio and video conferencing between a local conferencing system (i.e., local site or endpoint) and a remote conferencing system (i.e., remote site or endpoint). Audio/video conferencing application 502 controls local audio and video processing and establishes links with the remote site for transmitting and receiving audio and video over the ISDN or LAN network 110. Similarly, data conferencing application 504, also running on host processor 202, provides the top-level local control of data conferencing between the local and remote sites. Conferencing applications 502 and 504 communicate with the audio, video, and comm subsystems using conference manager 544, conferencing application programming interface (API) 506, LAN management interface (LMI) API 556, LMI manager 558, video API 508, comm API 510, and audio API 512. The functions of conferencing applications 502 and 504 and the APIs they use are described in further detail later in this specification.

Audio Processing

During conferencing, audio I/O hardware 410 of audio/comm (ISDN) board 206 digitizes analog audio signals received from microphone 104 and stores the resulting uncompressed digital audio to memory 404 via ISA bus interface 408. Audio task 538, running on DSP 406, controls the compression of the uncompressed audio and stores the resulting compressed audio back to memory 404.

Audio Processing for ISDN-Based Processing

For ISDN-based conferencing, comm task 540, also running on DSP 406, formats the locally-generated compressed audio for ISDN transmission and transmits the compressed ISDN-formatted audio to ISDN interface 402 for transmission to the remote site over ISDN network 110.

During ISDN-based conferencing, ISDN interface 402 also receives from ISDN network 110 compressed ISDN-formatted audio generated by the remote site and stores the compressed ISDN-formatted audio to memory 404. Comm task 540 then reconstructs the compressed audio format and stores the compressed audio back to memory 404. Audio task 538 controls the decompression of the compressed audio and stores the resulting decompressed audio back to memory 404. ISA bus interface then transmits the decompressed audio to audio I/O hardware 410, which digital-to-analog (D/A) converts the decompressed audio and transmits the resulting analog audio signals to speaker 108 for play.

Thus, for ISDN-based conferencing, audio capture/compression and decompression/playback are performed entirely within audio/comm (ISDN) board 206 without going through the host processor. As a result, audio is continuously played during an ISDN-based conferencing session regardless of what other applications are running on host processor 202.

Audio Processing for LAN-Based Processing

For LAN-based conferencing, audio task 538 passes the locally-generated compressed audio to the audio manager 520, which sends the compressed audio via comm API 510 to the comm manager 518 for transmission by the network stack 560 to the remote site via the LAN network 110.

During LAN-based conferencing, the network stack 560 also receives from LAN network 110 compressed LAN-formatted audio generated by the remote site and passes the compressed LAN-formatted audio to comm manager 518. Comm manager 518 then reconstructs the compressed audio format and passes the compressed audio via audio API 512 to audio manager 520, which stores the compressed audio into memory 404 of the audio/comm (ISDN) board 206 of FIG. 4. As in ISDN-based conferencing, audio task 538 controls the decompression of the compressed audio and stores the resulting decompressed audio back to memory 404. ISA bus interface then transmits the decompressed audio to audio I/O hardware 410, which digital-to-analog (D/A) converts the decompressed audio and transmits the resulting analog audio signals to speaker 108 for play.

Video Processing

Concurrent with the audio processing, video A/D converter 308 of video board 204 digitizes analog video signals received from camera 102 and transmits the resulting digitized video to video capture module 306. Video capture module 306 decodes the digitized video into YUV color components and delivers uncompressed digital video bitmaps to VRAM 304 via video bus 312. Video microcode 530, running on pixel processor 302, compresses the uncompressed video bitmaps and stores the resulting compressed video back to VRAM 304. ISA bus interface 310 then transmits via ISA bus 208 the compressed video to video/host interface 526 running on host processor 202.

Video/host interface 526 passes the compressed video to video manager 516 via video capture driver 522. Video manager 516 calls audio manager 520 using audio API 512 for synchronization information. Video manager 516 then time-stamps the video for synchronization with the audio. Video manager 516 passes the time-stamped compressed video to comm manager 518 via comm API 510.

Video Processing for ISDN-Based Conferencing

For ISDN-based conferencing, comm manager 518 passes the locally-generated compressed video through digital signal processing (DSP) interface 528 to ISA bus interface 408 of audio/comm (ISDN) board 206, which stores the compressed video to memory 404. Comm task 540 then formats the compressed video for ISDN transmission and transmits the ISDN-formatted compressed video to ISDN interface 402 for transmission to the remote site over ISDN network 110.

During ISDN-based conferencing, ISDN interface 402 also receives from ISDN network 110 ISDN-formatted compressed video generated by the remote site system and stores the ISDN-formatted compressed video to memory 404. Comm task 540 reconstructs the compressed video format and stores the resulting compressed video back to memory 404. ISA bus interface then transmits the compressed video to comm manager 518 via ISA bus 208 and DSP interface 528. Comm manager 518 passes the compressed video to video manager 516 via video API 508. Video manager 516 passes the compressed video to video decode driver 548 for decompression processing. Video decode driver 548 passes the decompressed video to video playback driver 550, which formats the decompressed video for transmission to the graphics device interface (GDI) (not shown) of the Microsoft.RTM. Windows.TM. operating system for eventual display in a video window on monitor 106.

Video Processing for LAN-Based Conferencing

For LAN-based conferencing, comm manager 518 formats the locally-generated compressed video for LAN transmission and transmits the LAN-formatted compressed video to the network stack 560 for transmission to the remote site over LAN network 110.

During LAN-based conferencing, the network stack 560 also receives from LAN network 110 LAN-formatted compressed video generated by the remote site system and passes the LAN-formatted compressed video to comm manager 518. Comm manager 518 then reconstructs the compressed video format and passes the compressed video via video API 508 to video manager 516. As in ISDN-based conferencing, video manager 516 passes the compressed video to video decode driver 548 for decompression processing. Video decode driver 548 passes the decompressed video to video playback driver 550, which formats the decompressed video for transmission to the graphics device interface (GDI) (not shown) of the Microsoft.RTM. Windows.TM. operating system for eventual display in a video window on monitor 106.

Data Processing

For data conferencing, concurrent with audio and video conferencing, data conferencing application 504 generates and passes data to comm manager 518 using conferencing API 506 and comm API 510.

Data Processing for ISDN-Based Conferencing

For ISDN-based conferencing, comm manager 518 passes the locally-generated data through board DSP interface 532 to ISA bus interface 408, which stores the data to memory 404. Comm task 540 formats the data for ISDN transmission and stores the ISDN-formatted data back to memory 404. ISDN interface 402 then transmits the ISDN-formatted data to the remote site over ISDN network 110.

During ISDN-based conferencing, ISDN interface 402 also receives from ISDN network 110 ISDN-formatted data generated by the remote site and stores the ISDN-formatted data to memory 404. Comm task 540 reconstructs the data format and stores the resulting data back to memory 404. ISA bus interface 408 then transmits the data to comm manager 518, via ISA bus 208 and DSP interface 528. Comm manager 518 passes the data to data conferencing application 504 using comm API 510 and conferencing API 506. Data conferencing application 504 processes the data and transmits the processed data to Microsoft.RTM. Windows.TM. GDI (not shown) for display in a data window on monitor 106.

Data Processing for LAN-Based Conferencing

For LAN-based conferencing, comm manager 518 formats the locally-generated data for LAN transmission and transmits the LAN-formatted data video to the network stack 560 for transmission to the remote site over LAN network 110.

During LAN-based conferencing, the network stack 560 also receives from LAN network 110 LAN-formatted data generated by the remote site system and passes the LAN-formatted data to comm manager 518. Comm manager 518 then reconstructs the data and passes the data to data conferencing application 504 using comm API 510 and conferencing API 506. As in ISDN-based conferencing, data conferencing application 504 processes the data and transmits the processed data to Microsoft.RTM. Windows.TM. GDI (not shown) for display in a data window on monitor 106.

Hardware Configuration for Conferencing System

LAN board 210 of FIG. 2 may be any suitable board for transmitting and receiving digital packets over a local (or wide) area network and is preferably an Intel.RTM. EtherExpress.TM. 16 card with appropriate control and network protocol software. Conferencing system 100 is capable of supporting LAN-based conferencing under different LAN transport standards (e.g., Novell IPX, Internet User Datagram Protocol (UDP), and/or NetBIOS standards). Furthermore, conferencing system 100 is capable of supporting LAN-based conferencing with different LAN products for a single LAN transport standard (e.g., LAN WorkPlace (LWPUDP) by Novell and FTPUDP by FTP Software, Inc., both of which conform to the LAN UDP standard). Thus, LAN board 210 corresponds to the LAN transports that are supported in conferencing system 100. Those skilled in the art will understand that more than one network stack 560 may be used to interface with a single LAN board 210.

Referring now to FIG. 6, there is shown a block diagram of the hardware configuration of audio/comm (ISDN) board 206 of FIG. 4. Referring now to FIGS. 30 and 31, there are shown diagrams of the architecture of the audio/comm (ISDN) board 206. Referring now to FIG. 32, there is shown a diagram of the audio/comm (ISDN) board environment. The description for the rest of this section is the same as the description for the section of the same name in U.S. patent application Ser. No. 08/157,694.

Software Architecture for Conferencing System

The software architecture of conferencing system 100 of FIGS. 2 and 5 has three layers of abstraction. A computer supported collaboration (CSC) infrastructure layer comprises the hardware (i.e., video board 204, audio/comm (ISDN) board 206, and LAN board 210) and host/board driver software (i.e., video/host interface 526, DSP interface 528, and network stack 560) to support video, audio, and comm, as well as the encode method for video (running on video board 204) and encode/decode methods for audio (running on audio/comm (ISDN) board 206). The capabilities of the CSC infrastructure are provided to the upper layer as a device driver interface (DDI).

A CSC system software layer provides services for instantiating and controlling the video and audio streams, synchronizing the two streams, and establishing and gracefully ending a call and associated communication channels. This functionality is provided in an application programming interface (API). This API comprises the extended audio and video interfaces and the communications APIs (i.e., conference manager 544, conferencing API (VCI) 506, LAN management interface (LMI) API 556, LMI manager 558, video API 508, video manager 516, video capture driver 522, video decode driver 548, video playback driver 550, comm API 510, comm manager 518, Wave API 514, Wave driver 524, PWave API 552, audio API 512, and audio manager 520).

A CSC applications layer brings CSC to the desktop. The CSC applications may include video annotation to video mail, video answering machine, audio/video/data conferencing (i.e., audio/video conferencing application 502 and data conferencing application 504), and group decision support systems.

Audio/video conferencing application 502 and data conferencing application 504 rely on conference manager 544 and conferencing API 506, which in turn rely upon video API 508, comm API 510, and audio API 512 to interface with video manager 516, comm manager 518, and audio manager 520, respectively. Comm API 510 and comm manager 518 provide a transport-independent interface (TII) that provides communications services to conferencing applications 502 and 504. The communications software of conferencing system 100 may be designed to support different transport mechanisms, such as ISDN, SW56, and LAN (e.g., SPX/IPX, TCP/IP, or NetBIOS). The TII isolates the conferencing applications from the underlying transport layer (i.e., transport-medium-specific DSP interface 528). The TII hides the network/connectivity specific operations. In conferencing system 100, the TII hides the ISDN and LAN layers. The DSP interface 528 is hidden in a datalink module (DLM). The LAN interface is hidden within a media dependent module (MDM). The TII provides services to the conferencing applications for opening communication channels (within the same session) and dynamically managing the bandwidth. The bandwidth is managed through a transmission priority scheme.

In an embodiment in which conferencing system 100 performs software video decoding, AVI capture driver 522 is implemented on top of video/host interface 526 (the video driver). In an alternative embodiment in which conferencing system 100 performs hardware video decoding, an AVI display driver is also implemented on top of video/host interface 526.

The software architecture of conferencing system 100 comprises three major subsystems: video, audio, and communication. The audio and video subsystems are decoupled and treated as "data types" (similar to text or graphics) with conventional operations like open, save, edit, and display. The video and audio services are available to the applications through video-management and audio-management extended interfaces, respectively.

Conferencing system 100 is implemented mostly in the C++ computer language using the Microsoft.RTM. Foundation Classes (MFC) with portions implemented in the C7.0 computer language.

Audio/Video Conferencing Application

The description for this section is the same as the description for the section of the same name in U.S. patent application Ser. No. 08/340,172, filed Nov. 15, 1994, now abandoned.

CMIF.LIB

The description for this section is the same as the description for the section of the same name in U.S. patent application Ser. No. 08/340,172, filed Nov. 15, 1994, now abandoned.

CCm

The description for this section is the same as the description for the section of the same name in U.S. patent application Ser. No. 08/340,172, filed Nov. 15, 1994, now abandoned.

Loading and Unloading

The description for this section is the same as the description for the section of the same name in U.S. patent application Ser. No. 08/340,172, filed Nov. 15, 1994, now abandoned.

Registering and Unregistering

The description for this section is the same as the description for the section of the same name in U.S. patent application Ser. No. 08/340,172, filed Nov. 15, 1994, now abandoned.

Call Support

The description for this section is the same as the description for the section of the same name in U.S. patent application Ser. No. 08/340,172, filed Nov. 15, 1994, now abandoned.

Channel Pair Support

The description for this section is the same as the description for the section of the same name in U.S. patent application Ser. No. 08/340,172, filed Nov. 15, 1994, now abandoned.

Stream Support

The description for this section is the same as the description for the section of the same name in U.S. patent application Ser. No. 08/340,172, filed Nov. 15, 1994, now abandoned.

CMDLL Callback

The description for this section is the same as the description for the section of the same name in U.S. patent application Ser. No. 08/340,172, filed Nov. 15, 1994, now abandoned.

NO VCI Support

The description for this section is the same as the description for the section of the same name in U.S. patent application Ser. No. 08/340,172, filed Nov. 15, 1994, now abandoned.

Miscellaneous

The description for this section is the same as the description for the section of the same name in U.S. patent application Ser. No. 08/340,172, filed Nov. 15, 1994, now abandoned.

CImageSize

The description for this section is the same as the description for the section of the same name in U.S. patent application Ser. No. 08/340,172, filed Nov. 15, 1994, now abandoned.

CImageState

The description for this section is the same as the description for the section of the same name in U.S. patent application Ser. No. 08/340,172, filed Nov. 15, 1994, now abandoned.

PSVIDEO.EXE

The description for this section is the same as the description for the section of the same name in U.S. patent application Ser. No. 08/340,172, filed Nov. 15, 1994, now abandoned.

Frame, View, and Image

The description for this section is the same as the description for the section of the same name in U.S. patent application Ser. No. 08/340,172, filed Nov. 15, 1994, now abandoned.

Class Descriptions

The description for this section is the same as the description for the section of the same name in U.S. patent application Ser. No. 08/340,172, filed Nov. 15, 1994, now abandoned.

CCyApp

The description for this section is the same as the description for the section of the same name in U.S. patent application Ser. No. 08/340,172, filed Nov. 15, 1994, now abandoned.

CCyFrameWnd

The description for this section is the same as the description for the section of the same name in U.S. patent application Ser. No. 08/340,172, filed Nov. 15, 1994, now abandoned.

CCyAppFrame

The description for this section is the same as the description for the section of the same name in U.S. patent application Ser. No. 08/340,172, filed Nov. 15, 1994, now abandoned.

CVideoFrame

The description for this section is the same as the description for the section of the same name in U.S. patent application Ser. No. 08/340,172, filed Nov. 15, 1994, now abandoned.

CVideoController

The description for this section is the same as the description for the section of the same name in U.S. patent application Ser. No. 08/340,172, filed Nov. 15, 1994, now abandoned.

Auto-Sizing of Video Windows

The description for this section is the same as the description for the section of the same name in U.S. patent application Ser. No. 08/340,172, filed Nov. 15, 1994, now abandoned.

Split and Combined Modes

The description for this section is the same as the description for the section of the same name in U.S. patent application Ser. No. 08/340,172, filed Nov. 15, 1994, now abandoned.

Control Channel Management

The description for this section is the same as the description for the section of the same name in U.S. patent application Ser. No. 08/340,172, filed Nov. 15, 1994, now abandoned.

Mute Message

The description for this section is the same as the description for the section of the same name in U.S. patent application Ser. No. 08/340,172, filed Nov. 15, 1994, now abandoned.

High-Quality Snapshot Message

The description for this section is the same as the description for the section of the same name in U.S. patent application Ser. No. 08/340,172, filed Nov. 15, 1994, now abandoned.

Application Launch

The description for this section is the same as the description for the section of the same name in U.S. patent application Ser. No. 08/340,172, filed Nov. 15, 1994, now abandoned.

Application Launch Response

The description for this section is the same as the description for the section of the same name in U.S. patent application Ser. No. 08/340,172, filed Nov. 15, 1994, now abandoned.

CChanPair

The description for this section is the same as the description for the section of the same name in U.S. patent application Ser. No. 08/340,172, filed Nov. 15, 1994, now abandoned.

Video View Class Relationships

The description for this section is the same as the description for the section of the same name in U.S. patent application Ser. No. 08/340,172, filed Nov. 15, 1994, now abandoned.

Handset Class Relationships

The description for this section is the same as the description for the section of the same name in U.S. patent application Ser. No. 08/340,172, filed Nov. 15, 1994, now abandoned.

Dialog Boxes

The description for this section is the same as the description for the section of the same name in U.S. patent application Ser. No. 08/340,172, filed Nov. 15, 1994, now abandoned.

Helper Classes

Dialog Helper

The description for this section is the same as the description for the section of the same name in U.S. patent application Ser. No. 08/340,172, filed Nov. 15, 1994, now abandoned.

Fast Bitmap Buttons

The description for this section is the same as the description for the section of the same name in U.S. patent application Ser. No. 08/340,172, filed Nov. 15, 1994, now abandoned.

Data Conferencing Application

The description for this section is the same as the description for the section of the same name in U.S. patent application Ser. No. 08/340,172, filed Nov. 15, 1994, now abandoned.

Conference Manager

The description for this section is the same as the description for the section of the same name in U.S. patent application Ser. No. 08/340,172, filed Nov. 15, 1994, now abandoned.

Conference Manager Overview

The description for this section is the same as the description for the section of the same name in U.S. patent application Ser. No. 08/340,172, filed Nov. 15, 1994, now abandoned.

Implementation Details

The description for this section is the same as the description for the section of the same name in U.S. patent application Ser. No. 08/340,172, filed Nov. 15, 1994, now abandoned. Additional information on the conference manager API is found in APPENDIX A of this specification.

Conference Application Installation

The description for this section is the same as the description for the section of the same name in U.S. patent application Ser. No. 08/340,172, filed Nov. 15, 1994, now abandoned.

Conference Application Registration

The description for this section is the same as the description for the section of the same name in U.S. patent application Ser. No. 08/340,172, filed Nov. 15, 1994, now abandoned.

VCI Call Handler Callback

The description for this section is the same as the description for the section of the same name in U.S. patent application Ser. No. 08/340,172, filed Nov. 15, 1994, now abandoned.

Channel Pair Establishment

The description for this section is the same as the description for the section of the same name in U.S. patent application Ser. No. 08/340,172, filed Nov. 15, 1994, now abandoned.

Critical Sections

The description for this section is the same as the description for the section of the same name in U.S. patent application Ser. No. 08/340,172, filed Nov. 15, 1994, now abandoned.

Call Notification and Caller ID

The description for this section is the same as the description for the section of the same name in U.S. patent application Ser. No. 08/340,172, filed Nov. 15, 1994, now abandoned.

Audible Call Progress

The description for this section is the same as the description for the section of the same name in U.S. patent application Ser. No. 08/340,172, filed Nov. 15, 1994, now abandoned.

On Demand Application Invocation

Referring now to FIG. 33, there is shown a flow diagram of the on-demand application invocation processing of conferencing system 100 of FIG. 5. The description for this section is the same as the description for the section of the same name in U.S. patent application Ser. No. 08/340,172, filed Nov. 15, 1994, now abandoned.

Managing Multiple Applications

Referring now to FIG. 34, there is shown a flow diagram of an example of the processing implemented within conferencing system 100 of FIG. 5 to manage two conferencing applications in a single conferencing session with a remote conferencing system. The description for this section is the same as the description for the section of the same name in U.S. patent application Ser. No. 08/340,172, filed Nov. 15, 1994, now abandoned.

Conferencing API

Referring now to FIG. 7, there is shown a block diagram of conference manager 544 and conferencing API 506 between conferencing applications 502 and 504, on one side, and comm API 508, LMI API 556, video API 510, and audio API 512, on the other side. The description for this section is the same as the description for the section of the same name in U.S. patent application Ser. No. 08/340,172, filed Nov. 15, 1994, now abandoned. Additional information on the conferencing API is found in APPENDIX B of this specification.

Interfacing with the Comm Subsystem

Conferencing API 506 supports the following comm services with the comm subsystem:

Comm initialization--initialize a session in the comm subsystem on which the call will be made.

Call establishment--place a call to start a conference.

Channel establishment--establish two comm channels for video conferencing control information, two comm channels for audio (incoming/outgoing), four comm channels for video (incoming data and control and outgoing data and control).

Call termination--hang up a call and close all active channels.

Comm Initialization/Uninitialization

Initialization of a session in the comm subsystem on which a call may be made by the user of conferencing system A of FIG. 1 and the user of conferencing system B of FIG. 1 is implemented as follows:

Conferencing APIs A and B call LMI.sub.-- AddLANTransport to initialize their LAN management interface (LMI) subsystems.

Conferencing APIs A and B receive a LMI.sub.-- ADDTRANS.sub.-- RESPONSE callback from the LMI subsystem.

Conferencing APIs A and B call BeginSession to initialize their comm subsystems.

Conferencing APIs A and B receive a SESS.sub.-- BEGIN callback from the comm subsystem.

Conferencing APIs A and B then notify the conferencing applications with a CFM.sub.-- INIT.sub.-- TRANSP.sub.-- NTFY callback.

Uninitialization of a session in the comm subsystem is implemented as follows:

Conferencing APIs A and B call LMI.sub.-- DeleteLANTransport to uninitialize their LAN management interface (LMI) subsystems.

Conferencing APIs A and B receive a LMI.sub.-- DELTRANS.sub.-- RESPONSE callback from the LMI subsystem.

Conferencing APIs A and B call EndSession to uninitialize their comm subsystems.

Conferencing APIs A and B receive a SESS.sub.-- CLOSED callback from the comm subsystem.

Conferencing APIs A and B then notify the conferencing applications with a CFM.sub.-- UNINIT TRANSP.sub.-- NTFY callback.

Call Establishment

Establishment of a call between the user of conferencing system A of FIG. 1 and the user of conferencing system B of FIG. 1 is implemented as follows:

Conferencing API A calls LMI.sub.-- RequestPermission to request permission to make the conference call from the management computer.

Conferencing API A receives a LMI.sub.-- PERM.sub.-- RESPONSE callback from the LMI subsystem. If permission is denied, conferencing API A notifies the conferencing application with a CFM.sub.-- REJECT.sub.-- NTFY callback. If permission is granted, establishment of the call is continued.

Conferencing API A calls LMI.sub.-- CallCommit to indicate to LMI that the call will be made.

Conferencing API A calls MakeConnection to dial conferencing API B's number.

Conferencing API B receives a CONN.sub.-- REQUESTED callback from the comm subsystem.

Conferencing API B calls LMI.sub.-- RequestPermission to request permission to accept the conference call from the management computer.

Conferencing API B receives a LMI.sub.-- PERM.sub.-- RESPONSE callback from the LMI subsystem. If permission is denied, conferencing API B rejects the call with RejectConnection, and notifies the conferencing application with a CFM.sub.-- DENIAL.sub.-- NTFY callback. If permission is granted, establishment of the call is continued.

Conferencing API B sends the call notification to the graphic user interface (GUI) with a CFM.sub.-- CALL.sub.-- NTFY callback; and, if user B accepts the call via the GUI, conferencing API B proceeds with the following steps.

Conferencing API B calls LMI.sub.-- CallCommit to indicate to LMI that the call will be accepted.

Conferencing API B calls AcceptConnection to accept the incoming call from conferencing API A.

Conferencing APIs A and B receive CONN.sub.-- ACCEPTED callback from the comm subsystem.

Conferencing API A calls OpenChannel to open its outgoing conferencing control channel.

Conferencing API B receives the CHAN.sub.-- REQUESTED callback for the incoming control channel and accepts it via AcceptChannel. Then conferencing API B calls OpenChannel to open its outgoing conferencing control channel.

Conferencing API A receives the CHAN.sub.-- ACCEPTED callback for its outgoing control channel and calls RegisterChanHandler to receive channel callbacks from the comm subsystem. Then conferencing API A receives the CHAN.sub.-- REQUESTED callback for the incoming control channel and accepts it via AcceptChannel.

Conferencing API B receives the CHAN.sub.-- ACCEPTED callback for its outgoing control channel and calls RegisterChanHandler to receive channel callbacks from the comm subsystem.

Conferencing API A sends a Login Request on the control channel, which conferencing API B receives.

Conferencing API B sends a Login Response on the control channel, which conferencing API A receives.

Conferencing APIs A and B negotiate conference capabilities between themselves. Capabilities that are negotiated include: negotiation protocol version, audio compression algorithm, video compression algorithm, video frame rate, video capture resolution, video bitrate, and data sharing capabilities.

Conferencing API A sends a Capabilities Request on the control channel, specifying conference requirements, which conferencing API B receives.

Conferencing API B sends a Capabilities Response on the control channel, accepting or modifying conference requirements, which conferencing API A receives.

When conferencing APIs A and B agree upon conference capabilities, the capabilities are saved and will be communicated to the application via the CFM.sub.-- ACCEPT.sub.-- NTFY callback.

Conferencing API A calls OpenChannel to open its outgoing audio channel.

Conferencing API B receives the CHAN.sub.-- REQUESTED callback for the incoming audio channel and accepts it via AcceptChannel.

Conferencing API A receives the CHAN.sub.-- ACCEPTED callback for the outgoing audio channel.

The last three steps are repeated for the video data channel and the video control channel.

Conferencing API B then turns around and repeats the above 4 steps (i.e., opens its outbound channels for audio/video data/video control).

Conferencing API A sends Participant Information on the control channel, which conferencing API B receives.

Conferencing API B sends Participant Information on the control channel, which conferencing API A receives.

Conferencing APIs A and B call LMI.sub.-- ConferenceCommit to indicate to LMI that the conference is in progress.

Conferencing APIs A and B then notify the conferencing applications with a CFM.sub.-- ACCEPT.sub.-- NTFY callback.

Channel Establishment

Video and audio channel establishment is implicitly done as part of call establishment, as described above, and need not be repeated here. For establishing other channels such as data conferencing, the conferencing API passes through the request to the comm manager, and sends the comm manager's callback to the user's channel manager.

Call Termination

Termination of a call between users A and B is implemented as follows (assuming user A hangs up):

Conferencing API A unlinks local/remote video/audio streams from the network.

Conferencing API A calls LMI.sub.-- ConferenceLeave to indicate to LMI that the conference is being closed.

Conferencing API A then calls the comm subsystem's CloseConnection.

The comm subsystem implicitly closes all channels, and sends CHAN.sub.-- CLOSED callbacks to the conferencing API A.

Conferencing API A closes its remote audio/video streams on receipt of the CHAN.sub.-- CLOSED callback for its inbound audio/video channels, respectively.

Conferencing API A then receives the CONN.sub.-- CLOSE.sub.-- RESP callback after the call is cleaned up completely. Conferencing API A notifies its conferencing application with a CFM.sub.-- HANGUP.sub.-- NTFY callback.

In the meantime, conferencing API B would have received the CHAN.sub.-- CLOSED callbacks from the comm subsystem for all the closed channels.

Conferencing API B closes its remote audio/video streams on receipt of the CHAN.sub.-- CLOSED callback for its inbound audio/video channels, respectively.

Conferencing API B unlinks its local audio/video streams from the network on receipt of the CHAN.sub.-- CLOSED callback for its outbound audio/video channels, respectively.

Conferencing API B then receives a CONN.sub.-- CLOSED callback from the comm subsystem.

Conferencing API B calls LMI.sub.-- ConferenceLeave to indicate to LMI that the conference is being closed.

Conferencing API B then notifies its conferencing application with a CFM.sub.-- HANGUP.sub.-- NTFY callback.

Interfacing with the Audio and Video Subsystems

The description for this section is the same as the description for the section of the same name in U.S. patent application Ser. No. 08/340,172, filed Nov. 15, 1994, now abandoned.

Capture/Monitor/Transmit Local Streams

The description for this section is the same as the description for the section of the same name in U.S. patent application Ser. No. 08/340,172, filed Nov. 15, 1994, now abandoned.

Receive/Play Remote Streams

The description for this section is the same as the descriptions for the section of the same name in U.S. patent application Ser. No. 08/340,172, filed Nov. 15, 1994, now abandoned.

Control Local/Remote Streams

The description for this section is the same as the description for the section of the same name in U.S. patent application Ser. No. 08/340,172, filed Nov. 15, 1994, now abandoned.

Snap an Image from Local Video Streams

Referring now to FIG. 8, there is shown a representation of the conferencing call finite state machine (FSM) for a conferencing session between a local conferencing system (i.e., caller) and a remote conferencing system (i.e., callee). Referring now to FIG. 9, there is shown a representation of the conferencing stream FSM for each conferencing system participating in a conferencing session. The description for this section is the same as the description for the section of the same name in U.S. patent application Ser. No. 08/340,172, filed Nov. 15, 1994, now abandoned. Differences include changes to the CF.sub.-- Init function and new functions CF.sub.-- Uninit, CF.sub.-- InitTransport, CF.sub.-- UninitTransport, and CF.sub.-- ChangeTransportMaxVideoBitrate, as follows:

CF.sub.-- Init Initializes the LAN management interface (LMI), audio and video subsystems, and initializes data structures required for conferencing.

CF.sub.-- Uninit Uninitializes the LMI, audio, and video subsystems. If a conference call is in progress, it is gracefully destroyed.

CF.sub.-- InitTransport Initializes a LAN or ISDN transport stack so that conference calls may be made or received on a particular transport type. The maximum video bitrate allowed on this transport is specified.

CF.sub.-- UninitTransport Uninitializes a transport stack, so calls may no longer be made or received on a particular transport type.

CF.sub.-- ChangeTransportMaxVideoBitrate Changes the maximum video bitrate allowed on a transport.

These functions are defined in further detail later in this specification in APPENDIX B.

In addition, conferencing API 506 supports the following additional messages returned to conferencing applications 502 and 504 from the video, comm, and audio subsystems in response to some of the above-listed functions:

CFM.sub.-- INIT.sub.-- TRANSP.sub.-- NTFY Indicates that transport stack initialization has completed successfully or unsuccessfully.

CFM.sub.-- UNINIT.sub.-- TRANSP.sub.-- NTFY Indicates that transport stack uninitialization has completed.

CFM.sub.-- UNFNIT.sub.-- NTFY Indicates that the conferencing API subsystem uninitialization has completed.

CFM.sub.-- DENIAL.sub.-- NTFY Indicates that a call request initiated from the remote site has been received, but the local site was denied permission to accept the call by the management computer.

CFM.sub.-- ERROR.sub.-- NTFY Indicates that an error has occurred in the comm subsystem.

CFM.sub.-- KILL.sub.-- NTFY Indicates that the management computer has demanded the conference call be terminated.

Video Subsystem

The video subsystem of conferencing system 100 of FIG. 5 comprises video API 508, video manager 516, video decode driver 548, video playback driver 550, video capture driver 522, and video/host interface 526 running on host processor 202 of FIG. 2 and video microcode 530 running on video board 204.

In an embodiment of the invention of U.S. patent application Ser. No. 08/157,694 (filed Nov. 24, 1993), the video subsystem encoded and decoded video according to a single video compression technique, that, for purposes of this patent application, may be referred to as the ISDN-rate video (IRV) technique. The video processing and video bitstream format described in defined in U.S. patent application Ser. No. 08/157,694, now U.S. Pat. No. 5,506,954, corresponded to that IRV technique.

The video subsystem of the present invention, however, is capable of encoding and decoding video according to more than one video compression technique. In one embodiment, the video system is capable of encoding and decoding video using both the IRV technique and a multi-rate video (MRV) technique. The following sections of this specification refer primarily to the IRV technique. The MRV technique is described in further detail in later sections of this specification starting with the section entitled "Compressed Multi-Rate Video Bitstream."

Video API

Referring now to FIG. 10, there is shown a representation of the video FSM for the local video stream and the remote video stream of a conferencing system during a conferencing session. The description for this section is the same as the description for the section of the same name in U.S. patent application Ser. No. 08/157,694, now U.S. Pat. No. 5,506,954. Additional information on the video API is found in APPENDIX C of this specification.

Video Manager

Referring now to FIG. 11, there is shown a block diagram of the software components of video manager (VM) 516 of FIG. 5. The description for this section is the same as the description for the section of the same name in U.S. patent application Ser. No. 08/157,694, now U.S. Pat. No. 5,506,954.

Capture/Playback Video Effects

The description for this section is the same as the description for the section of the same name in U.S. patent application Ser. No. 08/157,694, now U.S. Pat. No. 5,506,954.

Video Stream Restart

Referring now to FIG. 12, there is shown a representation of a sequence of N walking key frames. The description for this section is the same as the description for the section of the same name in U.S. patent application Ser. No. 08/157,694, now U.S. Pat. No. 5,506,954.

Audio/Video Synchronization

The description for this section is the same as the description for the section of the same name in U.S. patent application Ser. No. 08/340,172, filed Nov. 15, 1994, now abandoned.

Alternative Timestamp Driver

FIG. 35 represents the flow of bits between two remote high-resolution counters used to maintain clock values over a conferencing network. The description for this section is the same as the description for the section of the same name in U.S. patent application Ser. No. 08/340,172, filed Nov. 15, 1994, now abandoned.

Bit Rate Throttling

The description for this section is the same as the description for the section of the same name in U.S. patent application Ser. No. 08/340,172, filed Nov. 15, 1994, now abandoned.

Multiple Video Formats

The description for this section is the same as the description, for the section of the same name in U.S. patent application Ser. No. 08/340,172, filed Nov. 15, 1994, now abandoned.

Normal Display Resolution

The description for this section is the same as the description for the section of the same name in U.S. patent application Ser. No. 08/340,172, filed Nov. 15, 1994, now abandoned.

Quarter Display Resolution

The description for this section is the same as the description for the section of the same name in U.S. patent application Ser. No. 08/340,172, filed Nov. 15, 1994, now abandoned.

Video Frame Format/Capture Interface

The description for this section is the same as the description for the section of the same name in U.S. patent application Ser. No. 08/340,172, filed Nov. 15, 1994, now abandoned.

Playback Implementation

The description for this section is the same as the description for the section of the same name in U.S. patent application Ser. No. 08/340,172, filed Nov. 15, 1994, now abandoned.

Self-Calibration

The description for this section is the same as the description for the section of the same name in U.S. patent application Ser. No. 08/340,172, filed Nov. 15, 1994, now abandoned.

Measurement

The description for this section is the same as the description for the section of the same name in U.S. patent application Ser. No. 08/340,172, filed Nov. 15, 1994, now abandoned.

File-Based Capture (File Playback)

The description for this section is the same as the description for the section of the same name in U.S. patent application Ser. No. 08/340,172, filed Nov. 15, 1994, now abandoned.

Playback Statistics

The description for this section is the same as the description for the section of the same name in U.S. patent application Ser. No. 08/340,172, filed Nov. 15, 1994, now abandoned.

VCost Function

The description for this section is the same as the description for the section of the same name in U.S. patent application Ser. No. 08/340,172, filed Nov. 15, 1994, now abandoned.

VM DLL

The description for this section is the same as the description for the section of the same name in U.S. patent application Ser. No. 08/340,172, filed Nov. 15, 1994, now abandoned.

VCapt EXE

The description for this section is the same as the description for the section of the same name in U.S. patent application Ser. No. 08/340,172, filed Nov. 15, 1994, now abandoned.

VPlay EXE

The description for this section is the same as the description for the section of the same name in U.S. patent application Ser. No. 08/340,172, filed Nov. 15, 1994, now abandoned.

Palette Creation

The description for this section is the same as the description for the section of the same name in U.S. patent application Ser. No. 08/340,172, filed Nov. 15, 1994, now abandoned.

Extra RealizePalette Logic

The description for this section is the same as the description for the section of the same name in U.S. patent application Ser. No. 08/340,172, filed Nov. 15, 1994, now abandoned.

Netw DLL

The description for this section is the same as the description for the section of the same name in U.S. patent application Ser. No. 08/340,172, filed Nov. 15, 1994, now abandoned.

AVSync DLL

The description for this section is the same as the description for the section of the same name in U.S. patent application Ser. No. 08/340,172, filed Nov. 15, 1994, now abandoned.

Video Capture Driver

Video capture driver 522 of FIG. 5 follows driver specifications set forth in the Microsoft.RTM. Video for Windows.TM. (VfW) Developer Kit documentation. This documentation specifies a series of application program interfaces (APIs) to which video capture driver 522 responds. Microsoft.RTM. Video for Windows.TM. (VfW) is a Microsoft.RTM. extension to the Microsoft.RTM. Windows.TM. operating system. VfW provides a common framework to integrate audio and video into an application program. Video capture driver 522 extends the basic Microsoft.RTM. API definitions by providing nine "custom" APIs that provide direct control of enhancements to the standard VfW specification to enable and control bit rate throttling and local video monitoring. Video capture driver 522 captures images in the "raw" YVU9 format and compresses them using either the IRV or the MRV compression technique. Video capture driver 522 controls bit rate throttling and local video monitoring differently for IRV and MRV compression.

Bit rate throttling controls the bit rate of a transmitted video conference data stream. Bit rate throttling is based on the quality of the captured video image and the image capture frame rate. A high-quality image has more fine detail information than a low-quality image. A user of conferencing system 100 is able to vary the relative importance of image quality and frame capture rate with a custom capture driver API.

The data bandwidth capacity of the video conference communication channel is fixed. The amount of captured video data to be transmitted is variable, depending upon the amount of motion that is present in the video image. The capture driver is able to control the amount of data that is captured by changing the quality of the next captured video frame and by not capturing the next video frame (i.e., "dropping" the frame).

The image quality is determined on a frame-by-frame basis using the following equation: ##EQU1## Quality is the relative image quality of the next captured frame. A lower Quality number represents a lower image quality (less image detail). TargetSize is the desired size of a captured and compressed frame. TargetSize is based on a fixed, desired capture frame rate.

Normally, video capture driver 522 captures new video frames at a fixed, periodic rate which is set by the audio/video conference application program. Video capture driver 522 keeps a running total of the available communication channel bandwidth. When video capture driver 522 is ready to capture the next video frame, it first checks the available channel bandwidth and if there is insufficient bandwidth (due to a large, previously captured frame), then video capture driver 522 delays capturing the next video frame until sufficient bandwidth is available. Finally, the size of the captured video frame is subtracted from the available channel bandwidth total.

A user of conferencing system 100 may control the relationship between reduced image quality and dropped frames by setting image quality characteristics. For IRV compression, the user may set a minimum image quality value which controls the range of permitted image qualities, from a wide range down to a narrow range of only the best image qualities. For MRV compression, the user may set image quality using three parameters: motion estimation, spatial filtering, and temporal filtering. The effects of these parameters on image quality are discussed in U.S. patent application Ser. No. 08/235,955 (filed Apr. 28, 1994).

Bit rate throttling is implemented inside of the video capture driver and is controlled by the following VfW extension APIs:

CUSTOM.sub.-- SET.sub.-- DATA.sub.-- RATE Sets the data rate of the communications channel.

CUSTOM.sub.-- SET.sub.-- FPS Sets the desired capture frame rate.

CUSTOM.sub.-- SET.sub.-- QUAL.sub.-- PERCENT Sets the minimum image quality value (IRV only).

CUSTOM.sub.-- SET.sub.-- MOTION.sub.-- EST Enables or disables motion estimation (MRV only).

CUSTOM.sub.-- SET.sub.-- SPATIAL.sub.-- FILT Enables or disables spatial filtering (MRV only).

CUSTOM.sub.-- SET.sub.-- TEMPORAL.sub.-- FILT Sets the level of temporal filtering (MRV only).

The local video monitoring extension to VfW gives the video capture driver 522 the ability to output simultaneously both a compressed and a non-compressed image data stream to the application, while remaining fully compatible with the Microsoft.RTM. VfW interface specification. Without this capability, audio/video conferencing application 502 would have to decompress and display the image stream generated by the capture driver in order to provide local video monitoring, which would place an additional burden on the host processor and may decrease the frame update rate of the displayed image.

According to the VfW interface specification, the compressed image data is placed in an output buffer. When local video monitoring is active, an uncompressed copy of the same image frame is appended to the output buffer immediately following the compressed image data. The capture driver generates control information associated with the output buffer. This control information reflects only the compressed image block of the output buffer and does not indicate the presence of the uncompressed image block, making local video monitoring fully compatible with other VfW applications. A reserved, 32-bit data word in the VfW control information block indicates to a local video monitor aware application that there is a valid uncompressed video image block in the output buffer. The application program may then read and directly display the uncompressed video image block from the output buffer.

For the IRV technique, the uncompressed image data may be in either Device Independent Bitmap (DIB) or YUV9 format. For the MRV technique, the YVU9 format is used for the uncompressed image data. DIB format images are a fixed size, whereas YUV9 format images may be increased in size while retaining image quality. For both IRV and MRV techniques, the YUV9 images are converted into DIB format by the video display driver before they are displayed on the computer monitor.

The capture driver allows the uncompressed video image to be captured either normally or mirrored (reversed left to right). In normal mode, the local video monitoring image appears as it is viewed by a video camera--printing appears correctly in the displayed image. In mirrored mode, the local video monitoring image appears as if it were being viewed in a mirror.

The CUSTOM.sub.-- SET.sub.-- DIB.sub.-- CONTROL extension API controls the local video monitoring capabilities of the video capture driver.

Custom APIs for Video Capture Driver

The CUSTOM.sub.-- SET.sub.-- FPS message sets the frame rate for a video capture. This message is used while in streaming capture mode.

The CUSTOM.sub.-- SET.sub.-- KEY message informs the capture driver to produce one key frame as soon as possible. The capture driver will typically produce one delta frame before the key frame. Once the key frame has been encoded, delta frames will typically follow.

The CUSTOM.sub.-- SET.sub.-- DATA.sub.-- RATE message informs the capture driver to set an output data rate. This data rate value is in KBits per second and typically corresponds to the data rate of the communications channel over which the compressed video data will be transmitted.

The CUSTOM.sub.-- SETQUAL.sub.-- PERCENT message controls the relationship between reducing the image quality and dropping video frames when the IRV compressed video data stream size exceeds the data rate set by the CUSTOM.sub.-- SET.sub.-- DATA.sub.-- RATE message. For example, a CUSTOM.sub.-- SET.sub.-- QUAL.sub.-- PERCENT value of 0 means that the driver should reduce the image quality as much as possible before dropping frames and a value of 100 means that video frames should be dropped before the image quality is lowered. The CUSTOM.sub.-- SET.sub.-- QUAL.sub.-- PERCENT message is used only with IRV compression.

The CUSTOM.sub.-- SET.sub.-- DIB.sub.-- CONTROL message controls the uncompressed DIB or YUV9 format image output. With IRV compression, the uncompressed image may be in DIB format at either (80.times.60) or (160.times.120) pixel resolution or may be in YVU9 format at (160.times.120) resolution. With MRV compression, only the (160.times.120) YVU9 image format is supported. All images are available in either mirrored (reversed left to right) or a normal image. This API controls the following four parameters:

Uncompressed image enable/disable

Mirrored/normal image

The uncompressed image size

Image data format (DIB or YVU9)

The default condition is for the uncompressed image to be disabled. Once set, these control flags remain in effect until changed by another CUSTOM.sub.-- SET.sub.-- DIB.sub.-- CONTROL message. The uncompressed image data is appended to the video data buffer immediately following the compressed image data. The uncompressed DIB or YUV9 data have the bottom scan-line data first and the top scan-line data last in the buffer.

The CUSTOM.sub.-- SET.sub.-- VIDEO message controls the video demodulator CONTRAST, BRIGHTNESS, HUE (TINT), and SATURATION parameters. These video parameters are also set by the capture driver at initialization and via a video control dialog box.

The CUSTOM.sub.-- SET.sub.-- MOTION.sub.-- EST message allows MRV motion estimation to be enabled or disabled to improve image quality. This message is used only with MRV compression.

The CUSTOM.sub.-- SET.sub.-- SPATIAL.sub.-- FILT message allows MRV spatial filtering to be enabled or disabled to improve image quality. This message is used only with MRV compression.

The CUSTOM.sub.-- SET.sub.-- TEMPORAL.sub.-- FILT message allows the MRV temporal filter strength to be altered to improve image quality. This message is used only with MRV compression.

Video Microcode

The video microcode 530 of FIG. 5 running on video board 204 of FIG. 2 performs video compression. The preferred video compression techniques are disclosed in later sections of this specification starting with the section entitled "Compressed Video Bitstream."

Audio Subsystem

Referring now to FIG. 13, there is shown a block diagram of the architecture of the audio subsystem. The description for this section is the same as the description for the section of the same name in U.S. patent application Ser. No. 08/157,694, now U.S. Pat. No. 5,506,954. In addition, referring again to FIG. 13, if the network connection is over a LAN, then the audio task 538 on System A sends the packetized, time-stamped audio data to the commstub task 1308, which sends it to the audio manager 520 on the host processor 202. The audio manager 520 passes the data to TII 510 for delivery to the remote system. The audio data from System B is delivered by TII 510 to the audio manager 520 on System A (on the host). The audio manager 520 sends the packet to the commstub task 1308 which passes it on to the audio task 538.

Audio API

Referring now to FIG. 14, there is shown a representation of the audio FSM for the local audio stream and the remote audio stream of a conferencing system during a conferencing session. The description for this section is the same as the description for the section of the same name in U.S. patent application Ser. No. 08/157,694, now U.S. Pat. No. 5,506,954. Additional information on the audio API is found in APPENDIX D of this specification.

Audio Manager

The description for this section is the same as the description for the section of the same name in U.S. patent application Ser. No. 08/157,694, now U.S. Pat. No. 5,506,954.

Audio Manager Device Driver Interface

The description for this section is the same as the description for the section of the same name in U.S. patent application Ser. No. 08/157,694, except for the following. The expected messages (generated by a Microsoft.RTM. OpenDriver SDK call to installable device drivers) and the drivers response are as follows:

DRV.sub.-- LOAD Reads any configuration parameters associated with the driver. Allocates any memory required for execution. This call is only made the first time the driver is opened.

DRV.sub.-- ENABLE Ensures that an audio/comm board is installed and functional. For audio/comm board 206 of FIG. 2, this means the DSP interface 532 is accessible. This call is only made the first time the driver is opened.

DRV.sub.-- OPEN This call is made each time OpenDriver is called. The audio manager can be opened once for input, once for output (i.e., it supports one full duplex conversation), and any number of times for device capabilities query. This call allocates the per application data. This includes information such as the callback and the application instance data and buffers for transferring audio between the host and the audio board for LAN connections. If this is an input or output call, it starts the DSP audio task and sets up communication between host and DSP audio task (e.g. setup mail boxes, register callbacks). If this is the first open of an input or output stream, it starts the commstub task.

The installable device driver will respond to the close protocol messages defined by Microsoft.RTM.. The expected messages (generated by the Microsoft.RTM. SDK CloseDriver call to installable device drivers) and the drivers response are as follows:

This call sequence is symmetric with respect to the call sequence generated by OpenDriver. It has the same characteristics and behavior as the open sequence does. Namely, it receives one to three messages from the CloseDriver call dependent on the driver's state and it generates one callback per CloseDriver call. Three messages are received when the driver's final instance is being closed. Only the DRV.sub.-- CLOSE message is generated for other CloseDriver calls.

DRV.sub.-- CLOSE message closes the audio thread that corresponds to the audio stream indicated by HASTRM. The response to the close message is in response to a message sent back from the board indicating that the driver has closed. Therefore, this call is asynchronous.

AM.sub.-- LINKIN Message

The AM.sub.-- LINKIN message is sent to the driver whenever the audio manager function ALinkIn is called. Param1 is a pointer to the following structure:

ToLink contains a BOOL value that indicates whether the stream is being linked in or unlinked (TRUE is linked in and FALSE is unlinked). If no error is detected and ToLink is TRUE, the channel and the playback stream should be linked together. The driver calls TII to determine whether the transport associated with the channel is ISDN. If so, the driver calls TII to determine the ID of the channel on the board associated with the TII channel ID. It then sends the Audio Task the ALINKIN.sub.-- TMSG message with the board channel ID as a parameter. This causes the Audio Task to link up with the specified comm channel and begin playing incoming audio. If the transport associated with the channel is not ISDN, the driver prepares to receive data from the specified TII channel and send the data to the commstub task. It then sends the Audio Task the ALINKIN.sub.-- HOST.sub.-- TMSG. This causes the Audio Task to link up with the commstub task to receive the audio data and play it.

Breaking the link between the audio stream handle and the channel ID is done when the ToLink field is set to FALSE. The audio manager sends the ALINKIN.sub.-- TMSG to the task along with the channel ID. The Audio Task responds to this message by unlinking the specified channel ID (i.e., it does not play any more audio).

Errors that the host task will detect are as follows:

The channel ID does not represents a valid read stream.

The audio stream handle is already linked or unlinked (detected on host).

The audio stream handle is not a playback handle.

If those or any interface errors (e.g. message pending) are detected the callback associated with this stream is notified immediately. If no errors are detected, the ALINKIN.sub.-- TMSG or ALINKIN.sub.-- HOST.sub.-- TMSG is issued to the DSP interface and the message pending flag is set for this stream. Upon receiving the callback for this message, the callback associated with this stream is made, and finally the message pending flag is unset.

AM.sub.-- LINKOUT Message

The AM.sub.-- LINKOUT message is sent to the driver whenever the audio manager function ALinkOut is called. Param1 is a pointer to the following structure:

ToLink contains a BOOL value that indicates whether the stream is being linked out or unlinked (TRUE is linked out and FALSE is unlinked). If no error is detected and ToLink is TRUE, the channel and the audio in stream should be linked together. The driver calls TII to determine whether the transport associated with the channel is ISDN. If so, the driver calls TII to determine the ID of the channel on the board associated with the TII channel ID. It then sends the Audio Task the ALINKOUT.sub.-- TMSG message with the board channel ID as a parameter. This causes the Audio Task to link up with the specified comm channel and send it captured audio. If the transport associated with the channel is not ISDN, the driver prepares to receive data from the commstub task and send it to the specified TII channel. It then sends the Audio Task the ALINKOUT.sub.-- HOST.sub.-- TMSG. This causes the Audio Task to link up with the commstub task to send it captured audio data.

Breaking the link between the audio stream handle and the channel ID is done when ToLink field is set to FALSE. The audio manager will send the ALINKOUT.sub.-- TMSG to the task along with the channel ID. The Audio Task will respond to this message by unlinking the specified channel ID (i.e. it won't send any more audio).

Errors that the host task will detect are as follows:

The channel ID does not represents a valid write stream.

The audio stream handle is already linked or unlinked (detected on host).

The audio stream handle is not a audio in handle.

If those or any interface errors (e.g., message pending) are detected, the callback associated with this stream is notified immediately. If no errors are detected, the ALINKOUT.sub.-- TMSG or ALINKOUT.sub.-- HOST.sub.-- TMSG is issued to the DSP interface and the message pending flag is set for this stream. Upon receiving the callback for this message, the callback associated with this stream is made, and finally the message pending flag is unset.

Audio Manager Interface with the DSP Interface

The description for this section is the same as the description for the section of the same name in U.S. patent application Ser. No. 08/157,694, now U.S. Pat. No. 5,506,954.

Host Processor to Audio/Comm Board Messages

The description for this section is the same as the description for the section of the same name in U.S. patent application Ser. No. 08/157,694, except for the following:

ALINKIN.sub.-- TMSG: Connects/disconnects the audio task with a virtual circuit supported by the network task. The local and remote channel IDs (valid on the board) are passed to the audio task in the first two DWORDs of the dwArgs array. The flag specifying whether to link or unlink is passed in the third DWORD.

ALINKIN.sub.-- HOST.sub.-- TMSG: Connects/disconnects the audio task with the commstub task to receive audio to the host. The flag specifying whether to link or unlink is passed to the audio task in the third DWORD of the dwArgs array. The first two DWORDS are ignored.

ALINKOUT.sub.-- TMSG: Connects the audio task with a virtual circuit supported by the network task. The local and remote channel IDs (valid on the board) are passed to the audio task in the first two DWORDs of the dwArgs array. The flag specifying whether to link or unlink is passed in the third DWORD.

ALINKOUT.sub.-- HOST.sub.-- TMSG: Connects the audio task with a virtual circuit supported by the network task. The flag specifying whether to link or unlink is passed to the audio task in the third DWORD of the dwArgs array. The first two DWORDS are ignored.

Audio/Comm Board to Host Processor Messages

The description for this section is the same as the description for the section of the same name in U.S. patent application Ser. No. 08/157,694, now U.S. Pat. No. 5,506,954.

Wave Audio Implementation

The description for this section is the same as the description for the section of the same name in U.S. patent application Ser. No. 08/157,694, now U.S. Pat. No. 5,506,954.

Audio Subsystem Audio/Comm (ISDN) Board-Resident Implementation

The description for this section is the same as the description for the section of the same name in U.S. patent application Ser. No. 08/157,694, now U.S. Pat. No. 5,506,954. In addition, the audio task 538 of FIG. 13 connects with the commstub task 1308. This interface allows the audio task to exchange compressed data packets of audio samples with the host 202, which is responsible for delivering them to the remote system when the network is not ISDN (e.g., LAN). As the name implies, this task is a standin for the comm task. The interface is the same as that between the audio task 538 and the comm task 540.

Audio Task Interface with Host Device Driver

The description for this section is the same as the description for the section of the same name in U.S. patent application Ser. No. 08/157,694, now U.S. Pat. No. 5,506,954.

Audio Task Interface with Audio Hardware

Referring now to FIG. 15, there is shown a block diagram of interface between the audio task 538 and the audio hardware of audio/comm (ISDN) board 206 of FIG. 13. The description for this section is the same as the description for the section of the same name in U.S. patent application Ser. No. 08/157,694, now U.S. Pat. No. 5,506,954.

Timestamp Driver

The description for this section is the same as the description for the section of the same name in U.S. patent application Ser. No. 08/157,694, now U.S. Pat. No. 5,506,954.

(De)Compression Drivers

The description for this section is the same as the description for the section of the same name in U.S. patent application Ser. No. 08/157,694, now U.S. Pat. No. 5,506,954.

Mixer/Splitter Driver

The description for this section is the same as the description for the section of the same name in U.S. patent application Ser. No. 08/157,694, now U.S. Pat. No. 5,506,954.

Mixer Internal Operation

The description for this section is the same as the description for the section of the same name in U.S. patent application Ser. No. 08/157,694, now U.S. Pat. No. 5,506,954.

Echo Suppression Driver

The description for this section is the same as the description for the section of the same name in U.S. patent application Ser. No. 08/340,172, filed Nov. 15, 1994, now abandoned.

Spectral Equalization

In one embodiment of the present invention, the microphone 104 and speaker 108 of a conferencing node of FIG. 1 are part of a single earpiece component, such as an Enterprise.TM. headset sold by Plantronics. Because the microphone is located away from the mouth and in physical contact with the user's head near the ear, the audio signals may become distorted. These distortions may be due to reverberation signals that reflect off the user's cheek, sounds from the user's mouth that become out of phase at the microphone, and/or the directionality/loss of the higher frequencies. These distortions may combine with artifacts of the audio coder to degrade the quality of the audio portion of a conferencing session.

Digital filtering is applied to the audio signals to attempt to correct for the distortions that result from using a combined microphone/speaker earpiece. When using the Plantronics Enterprise.TM. microphone, the digital filter is implemented using a cascade of a second-order high-pass Chebyshev Type I Infinite Impulse Response filter and a sixth-order Infinite Impulse Response filter designed using the Steiglitz approximation, which produces a 3 dB bump at 2 kHz to enhance perception.

This digital filtering is implemented as part of the equalizer stackable driver 1514 in the capture side audio processing as shown in FIG. 15. The equalizer driver 1514 can be selectively enabled or disabled. When the user selects a combined earpiece headset, then the equalizer driver 1514 is enabled and each audio frame is digitally filtered before being passed to the next driver on the audio stack (i.e., echo/suppression stackable driver 1512 of FIG. 15). When the user selects another configuration of microphone and speaker (e.g., a speakerphone or a directional boom microphone headset), then the equalizer driver 1514 is disabled and each audio frame is passed on to the echo/suppression driver 1512 without any processing. The equalizer driver 1514 is implemented as a driver under the Spectron Microsystems SPOX.TM. operating system.

Audio Task Interface with Comm Task

Referring again to FIG. 13, the audio task 538 sends and receives audio packets from either the comm task 540 or the commstub task 1308, depending on whether the network connection is over ISDN or LAN. The interface the audio task uses in the same in either case. Throughout this section, references to comm task 540 also apply to commstub task 1308.

The interface between the audio task to the audio hardware is based on SPOX streams. Unfortunately, SPOX streams connect tasks to source and sink device drivers, not to each other. Audio data are contained within SPOX array objects and associated with streams. To avoid unnecessary buffer copies, array objects are passed back and forth between the comm and audio subsystems running on the audio/comm board using SPOX streams and a pipe driver. The actual pipe driver used will be based on a SPOX driver called NULLDEV. Like Spectron's version, this driver simply redirects buffers it receives as an IO.sub.-- SINK to the IO.sub.-- SOURCE stream; no buffer copying is performed. Unlike Spectron's pipe driver, however, NULLDEV does not block the receiving task if no buffers are available from the sending stream and discards buffers received from the IO.sub.-- SOURCE stream if no task has made the IO.sub.-- SINK stream connection to the driver. In addition, NULLDEV will not block or return errors to the sender. If no free buffers are available for exchange with the sender's live buffer, NULLDEV returns a previously queued live buffer. This action simulates a dropped packet condition.

Setup and teardown of these pipes will be managed by a message protocol between the comm task and audio task threads utilizing the existing TMB mailbox architecture built into the Mikado DSP interface. The interface assumes that the comm task or commstub task is running, a network connection has been established, and channel ID's (i.e., virtual circuit ID's) have been allocated to the audio subsystem by the conferencing API. The interface requires the comm task and commstub task each to make available to the audio threads the handle to its local mailbox TMB.sub.-- MYMBOX. This is the mailbox a task uses to receive messages from the host processor. The mailbox handle is copied to a global memory location and retrieved by the threads using the global data package discussed later in this specification. The audio task chooses which mailbox to use, and thus whether to communicate with the comm task or the commstub task, based on which message it receives from the host. ALINKOUT.sub.-- TMSG and ALINKIN.sub.-- TMSG cause it to use the comm task mailbox, and ALINKOUT.sub.-- HOST.sub.-- TMSG and ALINKIN.sub.-- HOST.sub.-- TMSG cause ti to use the commstub task mailbox. In the case of an ISDN connection, the audio task becomes the channel handler for the audio channels. Otherwise, the audio driver on the host becomes the channel handler.

Message Protocol

Referring now to FIG. 16, there is shown a block diagram of the interface between the audio task 538 and the comm task 540 of FIGS. 5 and 13. The description for this section is the same as the description for the section of the same name in U.S. patent application Ser. No. 08/157,694, which applies to conferencing over an ISDN connection. In addition, for a LAN connection, the processing is analogous as for the ISDN connection, with the following differences:

The commstub task replaces the comm task.

The ALINKOUT.sub.-- HOST.sub.-- TMSG message replaces the ALINKOUT.sub.-- TMSG message.

The ALINKIN.sub.-- HOST.sub.-- TMSG message replaces the ALINKIN.sub.-- TMSG message.

The commstub task sends buffers to and receives buffers from the host.

Global Data Package

The description for this section is the same as the description for the section of the same name in U.S. patent application Ser. No. 08/157,694, now U.S. Pat. No. 5,506,954.

NULLDEV Driver

The SPOX image for the audio/comm board contains a device driver that supports interprocess communication though the stream (SS) package. The number of distinct streams supported by NULLDEV is controlled by a defined constant NBRNULLDEVS in NULLDEV.H. NULLDEV supports three streams. One is used for the audio task capture thread to communicate with the comm task for ISDN connection. Another is used by the playback thread to communicate with the comm task. The third is for the audio capture task to communicate with the commstub task for LAN connection. The assignment of device names to tasks is done by the following three constants in ASTASK.H:

#define AS.sub.-- CAPTURE.sub.-- PIPE "/null"

#define AS.sub.-- PLAYBACK.sub.-- PIPE "/null2"

#define AS.sub.-- HOST.sub.-- CAPTURE.sub.-- PIPE "/null3"

Support for additional streams may be obtained by changing the NBRNULLDEVS constant and recompiling NULLDVR.C. The SPOX config file is also adjusted by adding additional device name strings to this section as follows:

The next device in the sequence has devid=3.

SS.sub.-- get() calls to NULLDEV receive an error if NULLDEV's ready queue is empty. It is possible to SS.sub.-- put() to a NULLDEV stream that has not been opened for SS.sub.-- get() on the other end. Data written to the stream in this case is discarded. In other words, input live buffers are simply appended to the free queue. SS.sub.-- put() never returns an error to the caller. If no buffers exist on the free queue for exchange with the incoming live buffer, NULLDEV removes the buffer at the head of the ready queue and returns it as the free buffer.

PWave Subsystem

The PWave subsystem provides high-priority playback of digital audio signals contained in Microsoft.RTM. standard Wave files.

PWave API

The description for this section is the same as the description for the section of the same name in U.S. patent application Ser. No. 08/340,172, filed Nov. 15, 1994, now abandoned.

High Priority Playback Task

The description for this section is the same as the description for the section of the same name in U.S. patent application Ser. No. 08/340,172, filed Nov. 15, 1994, now abandoned.

PWave Protocol

The description for this section is the same as the description for the section of the same name in U.S. patent application Ser. No. 08/340,172, filed Nov. 15, 1994, now abandoned.

Comm Subsystem

The communications (comm) subsystem of conferencing system 100 of FIG. 5 comprises:

Comm API 510, comm manager 518, DSP interface 528, and portions of the network stacks 560 running on host processor 202 of FIG. 2,

Portions of the network stacks 560 running on LAN board 210, and

Comm task 540 running on audio/comm (ISDN) board 206.

The comm subsystem provides connectivity functions to the conferencing application programs 502 and 504. It maintains and manages the session, connection, and the virtual channel states. All the connection control, as well as data communication are done through the communication subsystem.

Referring now to FIG. 17, there is shown a block diagram of the comm subsystem of conferencing system 100 of FIG. 5. The comm subsystem consists of the following layers that reside on host processor 202, the audio/comm (ISDN) board 206, and LAN board 210:

Transport independent interface 510 (TII.DLL),

Datalink module 1702 (DLM.DLL+KPDAPI.DLL, where KPDAPI.DLL is the back-end of the DLM which communicates with the DSP interface 528),

Reliable datalink module 1704 (RDLM.DLL),

Global dynamic loader 1706 (GDL.DLL),

Global dynamic loader executable 1708 (GDLE.EXE),

Control (D channel) 1710,

D channel driver 1712,

Data comm tasks 1714,

B channel drivers 1716,

LAN datalink module 1718 (DLMLAN.DLL),

The appropriate LAN media dependent modules 1720 (MDM.DLLs),

The appropriate comm stacks 560, and

The MDM helper task 1722 (MDMHELPR.DLL).

TII 510, DLM 1702, DSP interface 528, RDLM 1704, DLMLAN 1718, the MDMs 1720, portions of the comm stacks 560, MDMHELPR 1722, GDL 1706, and GDLE.EXE 1708 reside entirely on the host processor. Control (D channel) 1710, D channel driver 1712, data comm tasks 1714, and B channel drivers 1716 reside on audio/comm (ISDN) board 206. Portions of the comm stacks 560 reside on the LAN board 210.

The comm interface provides a "transport independent interface" for the conferencing applications. This means that the comm interface hides all the network dependent features of the conferencing system. For ISDN connections, conferencing system 100 uses the ISDN Basic Rate Interface (BRI) which provides 2*64 KBits/sec data (B) channels and one signaling (D) channel (2B+D). Conferencing system 100 also uses conventional LAN connections.

The comm subsystem provides an interface by which the conferencing applications can gain access to the communication hardware. The goal of the interface is to hide the implementation of the connectivity mechanism and provide an easy to use interface. This interface provides a very simple (yet functional) set of connection control features, as well as data communication features. The conferencing applications use virtual channels for data communication. Virtual channels are simplex, which means that two virtual channels are open for full duplex communication between peers. Each conferencing application opens its outgoing channel which is write-only. The incoming (read-only) channels are created by "accepting" an "open channel" request from the peer.

ISDN-Based Conferencing

Referring now to FIG. 18, there is shown a block diagram of the comm subsystem architecture for two conferencing systems 100 participating in a conferencing session over an ISDN connection. The comm subsystem provides an asynchronous interface between the audio/comm (ISDN) board 206 and the conferencing applications 502 and 504.

The comm subsystem provides all the software modules that manage the two ISDN B channels. The comm subsystem provides a multiple virtual channel interface for the B channels. Each virtual channel is associated with transmission priority. The data queued for the higher priority channels are transmitted before the data in the lower priority queues. The virtual channels are unidirectional. The conferencing applications open write-only channels. The conferencing applications acquire read-only channels as a result of accepting a open channel request from the peer. The DLM supports the virtual channel interface.

During an ISDN-based conferencing session, the comm subsystem software handles all the multiplexing and inverse multiplexing of virtual channels over the B channels. The number of available B channels (and the fact that there is more than one physical channel available) is not a concern to the application.

The comm subsystem provides the D channel signaling software to the audio/comm (ISDN) board. The comm subsystem is responsible for providing the ISDN B channel device drivers for the audio/comm (ISDN) board. The comm subsystem provides the ISDN D channel device drivers for the audio/comm (ISDN) board. The comm software is certifiable in North America (U.S.A., Canada) and Europe. The signaling software is compatible with NII, AT&T Custom, and Northern Telecom DMS-100.

LAN-Based Conferencing

For LAN-based conferencing, the comm subsystem provides an asynchronous interface between the LAN board 210 and the conferencing applications 502 and 504. The comm subsystem provides all the software modules that manage the LAN communication network 110. The comm subystem provides a multiple virtual channel interface for the LAN interconnecton between the conferencing machines. Each virtual channel is associated with a transmission priority. The data queued for the higher priority channels are transmitted before the data in the lower priority queues. The virtual channels are unidirectional. The conferencing applications open write-only channels. The conferencing applications acquire read-only channels as a result of accepting an open channel request from the peer. The DLMLAN modules supports the virtual channel interface.

During a LAN-based conferencing session, the comm subsystem handles all the multiplexing and inverse multiplexing of virtual channels over the typically singular LAN interconnection. The number of network `sockets` or connection points is not a concern to the application.

When the video conferencing connection is across the LAN, comm stack 506 receives the compress audio generated by the remote site and stores it to host memory. The appropriate LAN MDM 1720 of FIG. 17 and DLMLAN 1718 then reconstructs the compressed audio stream as the sequence of packets supplied by the audio manager on the remote site to that site's LAN comm subsystem. The comm manager 518 then passes the audio packets to the audio manager 520, which sends the packets to the audio task on audio/comm (ISDN) board 206 for playback.

qMUX MULTIPLE CHANNEL STREAMING MODULE

The description for this section is the same as the description for the section of the same name in U.S. patent application Ser. No. 08/157,694, now U.S. Pat. No. 5,506,954. In addition, for LAN-based conferencing, the LAN implementation of the DLM interface (DLMLAN) 1718 provides the same functionality on the LAN that DLM 1702 does for ISDN-based conferencing, i.e., virtual channels and transport independent message sizes. The DLMLAN implementation is supported on another abstraction layer, the media dependent modules (MDMs) 1720. The MDMs have a common MDM API and they implement the required functionality on top of an existing LAN protocol stack (e.g., IPX, TCP/IP) A single MDM helper task (MDMHELPR) 1722 assists the MDMs by generating threads of execution for callbacks and data transmission.

Comm API

The description for this section is the same as the description for the section of the same name in U.S. patent application Ser. No. 08/157,694, now U.S. Pat. No. 5,506,954. In addition, sessions and connections have associated addresses, represented by the TADDR structure. A TADDR consists of a transport type and up to 80 bytes of addressing information. The transport type specifies if this is an ISDN or LAN address. Referring again to FIG. 17, TII 510 determines which DLM will be servicing a given address by passing it to the Global Dynamic Loader (GDL) module 1706. GDL 1706 and its associated helper task GDLE 1708 load the appropriate module into memory and return all of the DLM entry points to TII 510. If this is a LAN address, the DLMLAN 1718 will then consult GDL 1706 in order to load the appropriate MDM 1720. DLMLAN 1718 receives back from GDL 1706 a list of the appropriate MDM entry points. GDL 1706 and GDLE 1708 determine the appropriate DLM and MDM to load by reading the file GDL.INI which is written when the product is installed. This file specifies the MDMs that are appropriate based on the configuration of the user's machine. Further description of the operations of global dynamic loaders and global dynamic loader executables is presented in U.S. patent application Ser. No. 08/133,612, now U.S. Pat. No. 5,410,698. Additional information on the comm API is found in APPENDIX E of this specification.

Automatic Transport Detection

Conferencing system 100 of FIG. 1 is capable of supporting conferencing over different types of transports (e.g., ISDN and LAN). Moreover, conferencing system 100 is capable of supporting LAN-based conferencing under different LAN transport standards (e.g., Novell IPX, Internet User Datagram Protocol (UDP), and/or NetBIOS standards). Further still, conferencing system 100 is capable of supporting LAN-based conferencing with different LAN products for a single LAN transport standard (e.g., LAN WorkPlace (LWPUDP) by Novell and FTPUDP by FTP Software, Inc., both of which conform to the LAN UDP standard).

In order for a particular conferencing system 100 to be able to exercise the full range of its conferencing options, it knows which of the supported transports are installed. Conferencing system 100 is able to determine automatically which supported transports are installed. This automatic transport detection may be implemented at install time (i.e., when conferencing system 100 is installed in a PC node) and/or at run time (i.e., when conferencing system 100 is ready to begin conferencing).

Although different LAN products that conform to the same transport standard will generate data with the same packet format, they may have different APIs for generating and interpreting those packets. Thus, automatic transport detection determines which transport products are installed as well as which transport types and standards are supported. Each different supported transport will typically have a corresponding media dependent module (MDM). A goal of automatic transport detection is to identify (and store in the GDL.INI file) the specific MDMs to be used to communicate with the specific network transport stacks that are supported and installed in conferencing system 100.

Install-Time Processing

Conferencing systems 100 may be configured to support conferencing over different sets of transports. For example, a particular conferencing system 100 may support conferencing over ISDN, Novell IPX, and UDP, but not NetBIOS. The supported transports are presented to the conferencing application 502 of conferencing system 100 as a list of numbers corresponding to the supported transports. Possible supported transports are identified as follows:

ISDN: 0

NetBIOS: 1

Novell IPX: 7

UDP: 8

For the conferencing system 100 in the example, a list of supported transports is (0, 7, 8).

Referring now to FIG. 42, there is shown a flow diagram of the processing by conferencing system 100 of FIGS. 5 and 17 during the automatic transport detection implemented at install time. The conferencing application selects the next (in this case, first) supported transport in the list of supported transports for that particular conferencing system 100 (step 4202 of FIG. 42).

Conferencing system 100 should have one or more possible corresponding MDMs for each supported transport, where there may be more than one MDM when there is more than one product for a particular transport type/standard. The conferencing application selects and loads the next (in this case, first) possible MDM for the currently selected transport (step 4204). The conferencing application calls the MDM.sub.-- BeginSession function to attempt to initialize the network transport stack (step 4206). A session may be defined as the use by an application of a specific transport to send and/or receive data packets on a particular network address. The conferencing application calls the MDM.sub.-- BeginSession function to request that a session be initiated by the local network stack.

If the attempt to begin a session is successful (step 4208), then the currently selected MDM is the MDM that corresponds to the network transport stack (for the currently selected transport) that is actually configured in conferencing system 100. In that case, the identity of the currently selected MDM is written to the GDL.INI file for the currently selected transport (step 4210). Processing then continues to step 4214.

If, however, the attempt to begin the session is unsuccessful (step 4208), then the currently selected MDM is not the correct MDM for the configured network transport stack. In that case, the conferencing application determines whether there are any other possible MDMs for the currently selected transport (step 4212). If so, then processing returns to step 4204 to select and load the next possible MDM and attempt to begin a session using it. If there are no more possible MDMs for the currently selected transport, then the conferencing application determines whether there are any more transports in the list of supported transports (step 4214). If so, then processing returns to step 4202 to repeat the processing for the next supported transport. Otherwise, install-time automatic transport detection is complete.

A result of automatic transport detection is a GDL.INI file that has, for each configured transport, the correct MDM to service that transport. The GDL.INI file is used by the conferencing application at run time to select the MDM to load and use for conferencing over a particular transport.

Run-Time Processing

Automatic transport detection is implemented at run time to determine which transports can be used for an impending conferencing session. Inputs to run-time automatic transport detection are the list of supported transports and the GDL.INI file that was generated by running automatic transport detection at install time. For each supported transport and using the corresponding MDM identified in the GDL.INI file, the conferencing application attempts to begin a session. If the attempt to begin the session is successful, then the conferencing application knows it can use that transport for the conferencing session.

Referring now to FIG. 43, there is shown a block diagram showing the network connections made by conferencing system 100 of FIGS. 5 and 17 during the automatic transport detection implemented at run time. The sequence shown in FIG. 43 may be enumerated as follows:

(A) Conferencing application 502 calls conferencing manager 544 to begin a session using a specified transport.

(B) Conferencing manager 544 passes begin session request to conferencing API 506.

(D) Comm API 510 calls GDLE.EXE 1708 to load the LAN data link manager (DLMLAN) 1718 corresponding to the specified transport.

(E) GDLE.EXE 1708 accesses the GDL.INI file 4302 to determine the file name for the DLMLAN 1718 that services the specified transport.

(F) GDLE.EXE 1708 loads the appropriate DLMLAN 1718 into memory and sends the corresponding entry points into DLMLAN 1718 back to comm API 510.

(G) Comm API 510 calls DLMLAN 1718 to begin a session.

(H) DLMLAN 1718 calls GDLE.EXE 1708 to load the media dependent module (MDM) corresponding to the specified transport.

(I) GDLE.EXE 1708 accesses the GDL.INI file 4302 to determine the file name for the MDM that services the specified transport.

(J) GDLE.EXE 1708 loads the appropriate MDM 1720 into memory and sends the corresponding entry points into MDM 1720 back to DLMLAN 1718.

(K) DLMLAN 1718 calls MDM 1720 to begin a session.

(L) MDM 1720 tries to communicate with the network stack for which it is designed to begin a session.

If the MDM's attempt to communicate with the network stack is successful, then that success is communicated from MDM 1720 to DLMLAN 1718 to comm API 510 to conferencing API 506 to conference manager 544 to the conferencing application 502. The conferencing application 502 then knows that it can use that transport for the impending conferencing session. Similarly, if the MDM's attempt to communicate with the network stack does not succeed, then that failure is communicated through the various levels to the conferencing application 502, which then knows that the conferencing session cannot proceed over that transport. In this latter case, as the MDM's attempt to begin the session fails, DLMLAN 1718 calls GDLE.EXE 1708 to unload MDM 1720, and, as the DLMLAN's attempt to begin the session fails, comm API 510 calls GDLE.EXE 1708 to unload DLMLAN 1718.

The scenario presented in FIG. 43 is repeated for each of the supported transports in the conferencing application's list of supported transports to determine all of the transports that are currently available for the conferencing session. When presenting the user with a list of possible callee addresses for the conferencing session (as part of a dialog box), the conferencing application 502 only lists addresses that correspond to transports that the conferencing application 502 has determined are available.

Transport-Independent Interface

The description for this section is the same as the description for the section of the same name in U.S. patent application Ser. No. 08/340,172, filed Nov. 15, 1994, now abandoned. These functions are defined in further detail later in this specification in Appendix E.

Message and Callback Parameters

The description for this section is the same as the description for the section of the same name in U.S. patent application Ser. No. 08/157,694, now abandoned.

Session Handler Messages

The description for this section is the same as the description for the section of the same name in U.S. patent application Ser. No. 08/340,172, filed Nov. 15, 1994, now abandoned.

Channel Manager Messages

The description for this section is the same as the description for the section of the same name in U.S. patent application Ser. No. 08/157,694, now U.S. Pat. No. 5,506,954.

Channel Handler Messages

The description for this section is the same as the description for the section of the same name in U.S. patent application Ser. No. 08/157,694, now U.S. Pat. No. 5,506,954.

iTone

The description for this section is the same as the description for the section of the same name in U.S. patent application Ser. No. 08/340,172, filed Nov. 15, 1994, now abandoned. Further description of the iTone string and its use may be found in U.S. patent application Ser. No. 08/305,206, filed Sep. 13, 1994, now U.S. Pat. No. 5,600,684.

Data Structures

Referring now to FIG. 19, there is shown a representation of the comm subsystem application finite state machine (FSM) for a conferencing session between a local conferencing system (i.e., local site or caller) and a remote conferencing system (i.e., remote site or callee). Referring now to FIG. 20, there is shown a representation of the comm subsystem connection FSM for a conferencing

session between a local site and a remote site. Referring now to FIG. 21, there is shown a representation of the comm subsystem control channel handshake FSM for a conferencing session between a local site and a remote site. Referring now to FIG. 22, there is shown a representation of the comm subsystem channel establishment FSM for a conferencing session between a local site and a remote site. Referring now to FIG. 23, there is shown a representation of the comm system processing for a typical conferencing session between a caller and a callee. The description for this section is the same as the description for the section of the same name in U.S. patent application Ser. No. 08/157,694.

Comm Manager

The comm manager 518 of FIG. 5 comprises the following dynamically linked libraries of FIG. 17:

Transport independent interface (TII) 510,

Reliable datalink module (RDLM.DLL) 1704,

Datalink module interface (DLM.DLL) 1702,

LAN datalink module interface (DLMLAN.DLL) 1718,

One or more media dependent modules (MDM.DLL) 1720,

Global dynamic loader (GDL.DLL) 1706,

Global dynamic loader executable (GDLE.EXE) 1708, and

MDM helper (MDMHELPR.DLL) 1722.

The DLM interface is used by the TII to access the services of the ISDN audio/comm board 206. The DLMLAN interface is used by the TII to access the services of the LAN board 210. Other modules (i.e., KPDAPI.DLL and DSP.DRV) function as the interface to the audio/comm board and have no other function (i.e., they provide means of communication between the host processor portion of the DLM and the audio/comm portion of the DLM. The host processor portion of the DLM (i.e., DLM.DLL) uses the DSP interface 528 of FIG. 5 (under Microsoft.RTM. Windows.TM. 3.x operating system) to communicate with the ISDN audio/comm board side portions.

The TII provides the ability to specify whether or not a virtual channel is reliable. For reliable channels, TII employs the RDLM to provide reliablility on a virtual channel. This feature is used to indicate that the audio and video virtual channels are unreliable, and the data virtual channel is reliable.

Referring again to FIG. 17, TII 510 is a dynamic link library (DLL) that implements the comm API. There is only a single instance of the TII library running on the host and it supports multiple transport media and multiple connections. At the bottom, the TII library makes DLM calls that are directed to the specific DLM capable of handling the address (transport) type in question.

A data link manager (e.g., DLM 1702, DLMLAN 1718) handles one or more transport types. A DLM provides:

Fragmentation and re-assembly of large messages,

Implementation of logical channels within a connection,

Prioritization of data across channels, and

In-order delivery of messages, with message boundaries preserved.

A DLM may directly interface to the transport media device (e.g., in the case of ISDN connections) or it may call the relevant media dependent module (MDM) (e.g., in the case of LAN connections) for services.

All transport media specific functionality is encapsulated into a media dependent module (MDM) 1720. There is one MDM per transport medium/protocol. Possible MDMs are NetBIOS, IPX, POTS Modems, and TAPI (Mikado PBX). If the underlying transport medium inherently supports multiple connections (e.g., NetBIOS), then the MDM should provide it to the upper layers. Some MDMs will provide a only single connection (e.g., a POTS Modem MDM that supports a single external modem). The MDM provides functionality for connection establishment and tear-down, and reliable data transfer over the connection(s). It does not have any knowledge of logical data channels. In conferencing system 100, each MDM is implemented as a DLL.

Additionally, there are two support modules. The link packet manager (LPM) 1724 creates, destroys, and allocates link packets for the communications stack. A link packet is data structure shared between the host-resident DLM and an MDM. Link packets allow for efficient transfer of data between the DLM and MDM. The global dynamic loader (GDL) 1706 is responsible for bringing DLMs and MDMs into the system as needed and for discarding them when they are no longer used.

Data Link Manager

Referring now to FIG. 29, there are shown diagrams indicating typical TII-DLM connection setup and teardown sequences. The description for this section is the same as the description for the section of the same name in U.S. patent application Ser. No. 08/340,172, filed Nov. 15, 1994, now abandoned. One difference is the event structure EVENTSTRUCT, which is extended to return the data block on CONN.sub.-- REQUESTED and is defined as follows:

Parameters:

EventType Specifies the type of event which triggered the callback.

Status Indicates the status of the event.

DlmId Unique ID of the DLM performing the callback.

MdmId Unique ID of the MDM which processed the event.

DlmSessionId Indicates the Session ID, assigned by DLM, on which this event occurred.

DlmConnId Indicates the Connection Id, assigned by DLM, on which this event occurred.

Token The token value was given in the call to initiate an action. When the callback notifies the user that the action is complete, the token is returned in this field.

Addr Specifies the LPTADDR of the caller.

Characteristics This field is a LPCONNCHR to the connection characteristics.

UserData Pointer to the data specified in the UserData parameter of the DLM.sub.-- MakeConnection call for this connection.

UserDataLen Number of valid bytes in the UserData block.

The UserData and UserDataLen fields are valid only for the CONN.sub.-- REQESTED callback and may be used for CallerID information. If the user did not specify any data, the UserData field is NULL and UserDataLen is 0.

Other differences are the DLM.sub.-- MakeConnection and DLM.sub.-- RejectConnection functions, which are defined as follows:

DLM MakeConnection

WORD DLM.sub.-- MakeConnection (DWORD DlmSessionId, LPCONNCHR Characteristics, DWORD Token, LPTADDR RemoteAddress, LPDWORD DLMConnID, LPVOID UserData, WORD UserDataLen);

Parameters:

DlmSessionID identifier returned in DLM.sub.-- BeginSession

Characteristics Desired characteristics of the connection. Passed uninterpreted to the lower layers.

Token Uninterpreted token returned to the upper layer in the response callback.

RemoteAddress Address on the remote site on which to make the connection.

DLMConnID Output parameter specifying the DLM connection ID that will be valid when this connection is established.

UserData Pointer to up to 64 bytes of user defined data that is to be transmitted to the remote site with the connection request.

UserDataLen Number of bytes in the UserData block. If more than 64 bytes are specified, the first 64 are transmitted.

DLM RejectConnection

WORD DLM.sub.-- RejectConnection (DWORD DlmConnId, DWORD CallReference, WORD ReasonCode);

Parameters:

DlmConnID Connection identifier returned in the CONN.sub.-- REQESTED callback.

CallReference Identifier returned previously in the CONN.sub.-- REQESTED callback.

ReasonCode Uninterpreted word that is transmitted to the remote site as the reason for the rejection.

The reason code will be returned in the Status field of the DLM event structure for the CONN.sub.-- REJECTED callback. If the remote user did not specify a reason, the Status will be 0.

Interfaces--Channel Management & Data Transfer

The description for this section is the same as the description for the section of the same name in U.S. patent application Ser. No. 08/157,694, now U.S. Pat. No. 5,506,954.

MDM Interface

The media dependent module (MDM) 1720 of FIG. 17 hides the network specifics from the layers above it on the communications stack. It is the only module that is affected by a change in the physical network. Media dependent modules are described in further detail in U.S. patent application Ser. No. 08/133,612, now U.S. Pat. No. 5,410,698. Additional information on the MDM API is found in APPENDIX F of this specification.

MDMHelpr

The MDM helper component (MDMHelpr) 1722 of FIG. 17 is a server of all system level services to its client MDMs. In order to centralize critical system timer resources, MDMHelpr 1722 task centralizes all access Microsoft.RTM. Windows.TM. messages, task time timers, multi-media timers and MDM messages. The first MDM starts the MDMHelpr task, which creates a ten-millisecond multimedia timer, a one-second task time timer, and a client registration service. Subsequently, each MDM registers as a client for the timers and message passing services. This centralizes the Microsoft.RTM. Windows.TM. system resources which are multiplexed within the helper eliminating redundancy as new MDMs are loaded. The MDMs utilize the helper's multimedia timer for both send and receive queue processing, the one-second task timer for packet retries and updating the log file, and the message queues for communication messages and control flow. As each MDM is closed, the clients are registered with the last MDM causing the shutdown of the helper task. Also, during abnormal termination, the helper catches the Microsoft.RTM. Windows.TM. close message and correctly cleans up its resources avoiding crashes by Microsoft.RTM. Windows.TM. from improper shutdown. Additional information on the MDMHelpr API is found in APPENDIX G of this specification.

Link Packet Manager

The link packet manager (LPM) 1724 of FIG. 17 maintains the pool of free link packets. Both DLM and MDM request link packets and, when they are finished with them, send them back to the LPM to be returned to the free pool. Since requests for link packets can occur at interrupt time, the LPM can not allocate packets on the fly. It allocates all of its free pool of packets when it is initialized and continues to re-use only those packets. Therefore, both DLM and MDM are able to handle the case that a link packet is not available. Link packet managers are described in further detail in U.S. patent application Ser. No. 08/133,612, now U.S. Pat. No. 5,410,698. Additional information on the LPM API is found in APPENDIX H of this specification.

Global Dynamic Loader

The global dynamic loader (GDL) 1706 of FIG. 17 is responsible for loading all necessary DLMs and MDMs into memory. The advantage over statically loading the libraries is that the communications stack need not be fixed when the application is started. The application may decide which network transport to use and consequently which MDM to load. Global dynamic loaders are described in further detail in U.S. patent application Ser. No. 08/133,612, now U.S. Pat. No. 5,410,698. Additional information on the GDL API is found in APPENDIX I of this specification.

DSP Interface

The description for this section is the same as the description for the section of the same name in U.S. patent application Ser. No. 08/157,694, now U.S. Pat. No. 5,506,954.

Comm Task

The description for this section is the same as the description for the section of the same name in U.S. patent application Ser. No. 08/157,694, now U.S. Pat. No. 5,506,954.

LAN Management Interface (LMI) Subsystem

The LAN management interface (LMI) module controls the communications between conferencing system 100 and a management computer to obtain LAN bandwidth allocations from the management computer for conferences. LMI consists of LMI API 556 and LMI manager 558 of FIG. 5. LMI maintains an internal interface to its own windows application (LMITRD.EXE) which provides a windows task thread to LMI. A management computer manages the allocation of bandwidth for conferences on a network. Management computers and how they manage bandwidth allocation for network conferences are described in greater detail in U.S. patent application Ser. No. 08/340,172, filed Nov. 16, 1994, now abandoned, entitled "Managing Bandwidth Over a Computer Network" of Robert Alexander Marshall, et al. Additional information on the LMI API is found in APPENDIX J of this specification.

Application-Level Protocols

The application-level protocols for conferencing system 100 of FIG. 5 are divided into those for the video, audio, and data streams.

Video Protocol

Referring now to FIG. 24, there is shown a representation of the structure of a video packet as sent to or received from the comm subsystem. The description for this section is the same as the description for the section of the same name in U.S. patent application Ser. No. 08/340,172, filed Nov. 15, 1994, now abandoned. In addition, conferencing system 100 is capable of encoding and decoding video signals in more than one bitstream format. Conferencing system 100 supports an ISDN rate video (IRV) bitstream format and a multi-rate video (MRV) bitstream format. These formats are described in the following sections.

Compressed ISDN Rate Video Bitstream

Referring now to FIG. 25, there is shown a representation of the compressed ISDN rate video (IRV) bitstream for conferencing system 100. The description for this section is the same as the description for the section entitled "Compressed Video Bitstream" in U.S. patent application Ser. No. 08/157,694, now U.S. Pat. No. 5,506,954.

Decoding Procedure for IRV Bitstream Format

The description for this section is the same as the description for the section entitled "Video Decoding Procedure" in U.S. patent application Ser. No. 08/157,694, now U.S. Pat. No. 5,506,954.

Intra/Inter Decision Rules

The description for this section is the same as the description for the section of the same name in U.S. patent application Ser. No. 08/157,694, now U.S. Pat. No. 5,506,954.

Post Reconstruction Loop Filtering

The description for this section is the same as the description for the section of the same name in U.S. patent application Ser. No. 08/157,694, now U.S. Pat. No. 5,506,954.

Adaptive Loop Filter Switching Criteria

The description for this section is the same as the description for the section of the same name in U.S. patent application Ser. No. 08/157,694, now U.S. Pat. No. 5,506,954.

Design of Quantization Tables

The description for this section is the same as the description for the section of the same name in U.S. patent application Ser. No. 08/157,694, now U.S. Pat. No. 5,506,954.

Adaptive Transform Coefficient Scanning

The description for this section is the same as the description for the section of the same name in U.S. patent application Ser. No. 08/157,694, now U.S. Pat. No. 5,506,954.

Spatially Adaptive Quantization

The description for this section is the same as the description for the section of the same name in U.S. patent application Ser. No. 08/157,694, now U.S. Pat. No. 5,506,954.

Fast Statistical Decode

The description for this section is the same as the description for the section of the same name in U.S. patent application Ser. No. 08/157,694, now U.S. Pat. No. 5,506,954.

Contrast, Brightness and Saturation Controls

The description for this section is the same as the description for the section of the same name in U.S. patent application Ser. No. 08/157,694, now U.S. Pat. No. 5,506,954.

Compressed Multi-Rate Video Bitstream

Encode and decode processing of one embodiment of the multi-rate video (MRV) bitstream are described in U.S. patent application Ser. No. 08/235,955, now U.S. Pat. No. 5,493,514.

Picture Layer

According to a current embodiment, the compressed data for one picture has the following format:

The format of the picture header is as follows:

where * indicates a field that starts on a byte boundary, ! indicates an optional field, and var indicates a variable number of bits. The meaning of these fields is as follows:

Picture start code A unique code that cannot occur anywhere else in the bit stream that identifies the start of a frame.

VersionNumber Identifies a version of the MRV video bit stream. The current version number is 1.

FrameNumber A counter that indicates to the decoder the receipt of successive frames. An encoder should increment this field by 1 for each frame it encodes (with wraparound, so that the FrameNumber after 127 is 0). A decoder can then determine by looking at this field if a frame has been "lost," except for the rare case in which a multiple of 128 frames in a row are lost.

SliceStructure Optionally specifies the slice structure of the picture. The binary value of this field has the following meanings:

If SliceStructure=0, information about slices is contained in the slice headers which appear later in the bitstream. If SliceStructure=1, 2, 3, or 4, the slice structure of the picture is specified here, as shown in the table above, and there are no slice headers in the bitstream for this picture.

PictureSize Identifies the size of the encoded picture. The binary value of this field has the following interpretation:

Reserved This field is reserved for future use, and is set to zero.

ExtraInfo This bit specifies whether there is extra information in the picture header.

QuantMatrices Specifies what quantization matrices the decoder should use for this frame. There are 32 quantization matrices to be defined: 16 for use in intra blocks and 16 for use in inter blocks. These 32 matrices are specified by the contents of two base matrices (one for intra, one for inter), five quantization parameters, and a flag (PowersOf2). The possible values of the QuantMatrices field specify what the base matrices, parameters, and PowersOf2 flag are, according to the following table:

In this table, "in QuantData" means that the given item is to be found in the QuantData field of the picture header. "From the past" means that the parameter values (which were set on a previous frame) are inherited from the past.

TempFiltStrength Specifies the strength of the temporal loop-filter. If this field is 00, the temporal filter is turned off for this picture.

UsesSLF Specifies whether this picture uses the spatial loop filter or not. The setting of this bit changes the meaning of macroblock types read from the bit stream.

UsesBlockMV If set, specifies that this picture may contain block-based (as opposed to macroblock-based) motion vectors. The setting of this bit changes the meaning of macroblock types read from the bit stream.

IntraFlag If set to 1, denotes that this picture is entirely intra coded.

PictureXsize and PictureYsize

Contain the picture sizes divided by 4. Since these are 8-bit fields, this means that MRV video can support picture sizes up to 1020.times.1020. These fields are present only if the PictureSize field=7.

ExtraInfoSize This field is present only if the Extralnfo bit is set earlier in the picture header. It specifies the number of bytes of extra information (not including this field) that is present in the picture header.

ExtraInfoData Extra information for private use by an encoder or decoder. An MRV video decoder should simply skip over any ExtraInfoData present in a bitstream.

QuantData If present (as indicated by the QuantMatrices field), contains a definition of the quantization parameters, the PowersOf2 flag, and possibly the two base matrices as well. If this field is present, its first four bytes are as follows:

If indicated by the QuantMatrices field, the definition of the two base matrices follows. Each matrix consists of 64 6-bit fields packed into 48 bytes. The inter base matrix is first, followed by the intra. Each matrix is stored in "raster-scan" order.

Slice Layer

The data for one MRV video picture consists of a series of data groups for one or more slices. A slice is a contiguous group of one or more rows of macroblocks in the picture. The slices present in the bit stream "covers" all three planes (Y, U, and V) of the picture exactly (i.e., the whole picture is coded and no part of the picture is encoded more than once). Some slices may contain very little data (for example, if all macroblocks in the slice are empty), but they are all present in the data stream. Each slice in a picture is specified by start and size fields specifying the row and plane where the slice starts and its size (i.e., how many rows of macroblocks the slice contains). A slice is required to be confined to a single plane. The data for one slice consists of three sections, as follows:

The format of the slice header is as follows:

where:

Slice start code Identifies the start of a slice. This unique code cannot occur elsewhere in the bit stream.

SliceStart Specifies where the slice starts in the picture. The units of SliceStart are interpreted as follows: Take the rows of macroblocks in each of the Y, U, and V planes and arrange them in scan-line order, then concatenate the rows for the planes into one long list. The value of SliceStart is the (zero-based) index into this list. For example, in a (160.times.120) picture, there are 8 rows of macroblocks in the Y plane and 2 rows in each of the U and V planes. So SliceStart would take on a value of 0 to 11, where 0 represents the top row of macroblocks in the Y plane, 7 is the bottom row of the Y plane, 8 is the top row of the U plane, etc.

SliceSize Specifies the size of the slice in rows of macroblocks. A slice is confined to a single plane, but is allowed to start and end on any row. A value of SliceSize which describes a slice extending past the end of the plane is illegal. Slices in a picture need not appear in the bit stream "in order". For example, a picture could have 4 slices, in the following order: U plane, top half of Y plane, V plane, bottom half of Y plane.

Reserved A 2-bit field reserved for future use (0 for now).

Following the slice header is the Huffman-encoded macroblock and block data for one slice. The macroblock data always starts on a byte boundary, but the block data need not, as it is simply concatenated to the end of the macroblock data.

Macroblock Layer

The macroblock data describes the structure of each macroblock in a slice. The macroblock data consists of a series of "records" of the form:

followed by the "separator" symbol 00011111, which separates the macroblock data from the block data for the slice. All of the symbols in the above record are encoded using the Macroblock Huffman table. The meaning of each of these fields is as follows:

The empty field gives information about how many empty macroblocks there are between this non-empty macroblock and the previous one. There may be more than one empty value, to indicate a long run of empty macroblocks.

The type field actually contains several bits of information about the macroblock. The decoded Huffman value is used as an index into one of the following four tables. The table to use is determined (on a per-picture basis) by the settings of the UsesSLF and UsesBlockMV bits in the picture header.

The bits in these tables have the following meaning:

Intra Says whether this macroblock is intra or not.

NewQ Says whether a quantization index (Qvalue) is present in this record.

MV Says whether a motion vector is present in this record.

Cbp Says whether a Cbp (coded block pattern) value is present in this record.

SLF Says whether the spatial loop filter is to be used for this macroblock.

BlockMV Says whether this record contains four motion vectors (one for each block) or one.

Following the type field are the QValue, MV, and cbp fields, which are present only if indicated by the corresponding bit in the type.

QValue is the Huffman-encoded differential quantization value. The quantization value for this macroblock (thisQ) is calculated from the previous macroblock's value (prevQ) as follows:

where tosigned() is a function which converts from an unsigned number to a signed number. The resulting thisQ value is a 4-bit value that specifies which of 16 quantization matrices to use. The value of prevQ is initialized to 8 at the start of each slice.

If MV=1, there is either one (if BlockMV=0) or four (if BlockMV=1) motion vectors present in this record. Each motion vector consists of two separate values, one for the x component and one for the y component. For both x and y, the actual vector component is calculated as:

In these equations, prevMV is the motion vector of the previous block or macroblock, depending on whether this macroblock has one or four motion vectors, and whether the previous macroblock had one or four motion vectors, as follows:

MVn corresponds to block number n in the macroblock (according to the numbering shown in Macroblock and Block Structure). At the start of each row of macroblocks, the x and y components of prevMV are reset to zero. prevMV refers to the immediately-preceding macroblock. In particular, this means that if the preceding macroblock is empty, prevMV=0. A positive motion vector x component means that the prediction block in the previous picture is to the right of the block in the current picture. A positive y component means that the prediction block is below the block in the current picture. The x and y components are values in the range -21 to +21.

The cbp value in the record specifies which blocks in the macroblock are empty (i.e., have no coded transform coefficients). The cbp value is obtained by taking the Huffman-decoded value (which will be in the range 0-14), and indexing into the following table:

cbp is never zero, since in that case the Cbp bit in the type field would be set to 0, and cbp would not be present in the record. The 4-bit cbp value specifies the emptiness of the 4 blocks in the macroblock, with bit=1 meaning "non-empty". The LSB of cbp applies to block #1, the next bit to block #2, etc. There are also two special cases:

(1) If the macroblock type says "intra," cbp is not present but is implied to be 15.

(2) If the macroblock type says "non-intra" and "cbp not present," cbp is implied to be 0.

Block Layer

The block data contains data for each of the coded (non-empty, non-phantom) blocks in a slice. Blocks are coded in macroblock scan order, and within each macroblock in counterclockwise order starting at the top left corner. The block data is terminated by a string of 14 zero bits, which is the first unused code in the block Huffman table. There are two cases:

(1) If the block data is followed by a slice header for another slice, the block data is padded (with zero bits) to the next byte boundary. This can be followed by a string of or more "padding bytes" of zeros, followed by the slice header for the next slice. Regardless of the number of padding bits or bytes, the zero bits which begin the next slice header guarantee that the block data is terminated by a string of 14 zero bits.

(2) If the block data is not followed by a slice header (either because the picture header was used to specify the slice structure, or because this is the last slice of the picture), then the encoder explicitly inserts the 14 zero bits. Then, the block data is padded with zero bits to the next byte boundary. If another (header-less) slice follows, its macroblock data follow immediately and padding bytes are not allowed.

In both cases, the macroblock data for a slice starts on a byte boundary. The general format of the block data is as follows:

and the format of each block's data is as follows:

The basic unit of block data is the "run/value pair" or run/val pair for short. Each run/val pair represents one non-zero FST frequency-domain coefficient. The commonly-occurring run/val pairs are encoded with a single code; others are encoded as ESC followed by the explicit run and value.

Audio Protocol

Referring now to FIG. 26, there is shown a representation of a compressed audio packet for conferencing system 100. The description for this section is the same as the description for the section of the same name in U.S. patent application Ser. No. 08/340,172, filed Nov. 15, 1994, now abandoned.

Compressed Audio Bitstream

The description for this section is the same as the description for the section of the same name in U.S. patent application Ser. No. 08/340,172, filed Nov. 15, 1994, now abandoned.

Data Protocol

The description for this section is the same as the description for the section of the same name in U.S. Patent Application filed Nov. 15, 1994. Data conferencing application 504 is described in greater detail in U.S. patent application Ser. No. 08/137,319 (filed Oct. 14, 1993), now U.S. Pat. No. 5,452,299 and in U.S. patent application Ser. No. 08/170,146 (filed Dec. 20, 1993), now U.S. Pat. No. 5,581,702.

Communication-Level Protocols

The description for this section is the same as the description for the section of the same name in U.S. patent application Ser. No. 08/157,694, now U.S. Pat. No. 5,506,954.

Reliable Transport Comm Protocols

Referring now to FIG. 27, there is shown a representation of the reliable transport comm packet structure. The description for this section is the same as the description for the section of the same name in U.S. patent application Ser. No. 08/157,694, now U.S. Pat. No. 5,506,954.

Unreliable Transport Comm Protocols

Referring now to FIG. 28, there is shown a representation of the unreliable transport comm packet structure. The description for this section is the same as the description for the section of the same name in U.S. patent application Ser. No. 08/340,172, filed Nov. 15, 1994, now abandoned.

DLMLAN Protocols

The DLMLAN 1718 of FIG. 17 fragments the messages from TII 510 into packets to be transmitted to the network. As shown in FIG. 44, each packet contains the following header:

Following the DLM header is an implementation defined number of data bytes.

MDM Protocols

For LAN transmissions, the MDMs 1720 of FIG. 17 wrap the DLM packet with the following header, as shown in FIG. 45:

Following the header is an implementation defined number of bytes containing the data of the packet.

MDM Protocol for Connection Setup and Tear Down

The MDMs place logical connections on top of typically connectionless network protocols. Therefore, the MDMs exchange messages which build the MDM logical connection. There are four messages exchanged by the MDMs: connection request, connection acknowledgment, connection close, and clear to send.

Connection Request

The connection request message (Open) requests the creation of connection between two MDMs. This message requests the target MDM to respond using its listening connection to begin the connection accept and establish sequence. The response expected is either an ACK or NAK message. If neither is received in one second from transmitting the Open, another Open is sent. If no response is received by the requester within the timeout period, then the Open request is cancelled. The connection request is sent as the result of a call to MDM.sub.-- MakeConnection. It has the following format:

Immediately following source address (byte offset: 12 for IPX, 16 for UDP, 26 for NetBIOS) is the size of the user data block following the message (format: WORD). Immediately following the user data block size (byte offset: 14 for IPX, 18 for UDP, 28 for NetBIOS) is the user data block (format: uninterpreted block of up to 64 bytes of data sent to the MDM in MDM.sub.-- MakeConnection), which may be used for CallerID information.

Connection Acknowledgment

The connection acknowledgment message (ACK) is the response to the Open request, if the receiver has a connection listening and expects to establish a connection. The acknowledgement of the connection request has the following format:

Any remaining bytes of the packet are 0.

Connection Close

The connection close message (NAK) is the response to the connection request, if the receiver has no connection listening or does not expect to establish a connection. NAK is the request by either side of an accepted or established connection to close or destroy the connection. Currently, there is no response to this request. The close or reject message has the following format:

Any remaining bytes of the packet are 0.

Clear To Send

The clear to send message (CTS) is the request by the receiver of the connection request message that the connection has been accepted and that the connection is ready to receive data messages. The response expected is either CTS, NAK, or a connection data message. If no response is received by the requestor in one second from transmitting the CTS request, another CTS is sent. If no response is received by the requestor within the timeout period, the CTS request is cancelled and the connection is closed. The response to the CTS request if the receiver has requested a connection be opened and expects to establish a connection. A clear to send message has the following format:

Any remaining bytes of the packet are 0.

Referring now to FIG. 46, there is shown a representation of the connection messages for a typical conferencing session from the perspective of the MDMs on the local and remote nodes.

Feature and Capability Negotiation

Conferencing Management

The description for this section is the same as the description for the section of the same name in U.S. patent application Ser. No. 08/340,172, filed Nov. 15, 1994, now abandoned.

Connection Management

Connection and Channel Setup

The description for this section is the same as the description for the section of the same name in U.S. patent application Ser. No. 08/340,172, filed Nov. 15, 1994, now abandoned.

Connection Shutdown and Error Handling

The description for this section is the same as the description for the section of the same name in U.S. patent application Ser. No. 08/340,172, filed Nov. 15, 1994, now abandoned.

Conference Login

The description for this section is the same as the description for the section of the same name in U.S. patent application Ser. No. 08/340,172, filed Nov. 15, 1994, now abandoned.

Capability Negotiation

The description for this section is the same as the description for the section of the same name in U.S. patent application Ser. No. 08/340,172, filed Nov. 15, 1994, now abandoned.

Capabilities Structure

The description for this section is the same as the description for the section of the same name in U.S. patent application Ser. No. 08/340,172, filed Nov. 15, 1994, now abandoned.

Requests and Responses

The description for this section is the same as the description for the section of the same name in U.S. patent application Ser. No. 08/340,172, filed Nov. 15, 1994, now abandoned.

CallerCapRequest

The description for this section is the same as the description for the section of the same name in U.S. patent application Ser. No. 08/340,172, filed Nov. 15, 1994, now abandoned.

CallerCapCancel

The description for this section is the same as the description for the section of the same name in U.S. patent application Ser. No. 08/340,172, filed Nov. 15, 1994, now abandoned.

CalleeCapAccept

The description for this section is the same as the description for the section of the same name in U.S. patent application Ser. No. 08/340,172, filed Nov. 15, 1994, now abandoned.

CalleeCapReject

The description for this section is the same as the description for the section of the same name in U.S. patent application Ser. No. 08/340,172, filed Nov. 15, 1994, now abandoned.

CalleeCapResponse

The description for this section is the same as the description for the section of the same name in U.S. patent application Ser. No. 08/340,172, filed Nov. 15, 1994, now abandoned.

Video Negotiation

Conferencing system 100 is able to support different modes of video processing, where each different mode is defined by a set of four parameters:

(1) Bitstream format: the format of the compressed video bitstream (may also imply the techniques used to code and decode the compressed video stream).

(2) Frame rate: the maximum number of video frames per second.

(3) Frame resolutions: the dimensions of the video frames in number of pixels wide and number of pixels high.

(4) Bit rate: the maximum bit rate for the compressed video bitstream.

In a possible embodiment of the present invention, these four parameters are independent and theoretically limitless in variety. In another embodiment, however, the parameters are both limited and dependent. As to the limits to the parameters:

(1) The possible bitstream formats are limited to the so-called IRV bitstream (as defined in FIG. 25) and the so-called MRV bitstream (as defined in this specification in the section entitled "Compressed Multi-Rate Video Bitstream").

(2) The possible maximum frame rates are limited to integer numbers of frames per second from 1 to 30 fps, inclusive.

(3) The possible frame resolutions are limited to (160.times.120) and (320.times.240).

(4) The possible maximum bit rates are limited to 200 kilobits per second (kbps), 90 kbps, and 84 kbps.

As to the interdependency of the parameters, the allowable video modes are limited to a specified set of supported combinations of the four parameters. The set of supported combinations are enumerated in a table in the video subsystem, which identifies the relative preferences for the combinations. For example, a possible table may contain the following supported video mode:

Two nodes that want to have a video conference will negotiate with each other to select a video processing mode (i.e., a specific combination of the four parameters) for the conferencing session. In one mechanism for conducting video negotiation, each step in the video negotiation involves one node sending to the other node a capabilities data structure that encodes a particular set of video capabilities The data structure contains a 32-bit field encoding the capabilities in separate bits, where each bit is a yes/no value for a different combination such as those listed in the previous table. For example, the capabilities for all four video mode choices listed in the previous table may be encoded as follows:

In an alternative embodiment of the present invention, the mechanism for conducting video negotiation between two conferencing systems 100 of FIG. 1 assumes that the four video parameters are in fact independent. Under this mechanism, each step in the video negotiation involves one node sending to the other node a capabilities data structure that encodes a particular set of video capabilities. The data structure contains a 32-bit field, such that:

One byte encodes bitstream formats, where each bit is a yes/no value for a different bitstream format.

One byte encodes a maximum frame rate as an 8-bit value.

One byte encodes frame resolutions, where each bit is a yes/no value for a different frame resolution.

One byte encodes a maximum bit rate as an 8-bit value were the value of the LSB is 2.

For example, the capabilities for all four video mode choices listed above may be encoded as follows:

a Bitstream format byte=(00000011), where the LSB indicates that the IRV bitstream is supported and the next LSB indicates that the MRV bitstream is supported,

Frame rate byte=(00001111) corresponding to 15 fps.

Frame resolution byte=(00000001), where the LSB indicates that the (160.times.120) frame resolution is supported and the next LSB indicates that the (320.times.240) frame resolution is not supported.

Bit rate byte=(01011111) corresponding to 200 kbps.

In yet another alternative embodiment, video negotiation is based on combinations of bitstream format, maximum frame rate, and frame resolution, where the maximum bit rate is externally specified.

Referring now to FIG. 47, there is shown a flow diagram of the video negotiation processing between two conferencing systems 100 (i.e., node A and node B) of FIG. 1. Node A sends a negotiation proposal to node B, where the proposal contains a set of capabilities encoded in the 32-bit capabilities data structure as described above (step 4702). If the proposal corresponds to a unique video mode (i.e., the proposed capabilities include only one bitstream format and only one frame resolution) and if that proposed video mode is acceptable to node B (step 4704), then the video negotiations have successfully chosen a video mode acceptable to both nodes and node B sends the same proposal back to node A to accept the video mode for video conferencing (step 4706).

If, however, the proposal is not a unique video mode (e.g., there is more than one bitstream format and/or frame resolution) or if the proposal is not acceptable to node B (e.g., frame rate and/or bit rate is too high) (step 4704), node B determines whether it can make a counter proposal to node A (step 4708). A counter proposal is a subset of the capabilities contained in the previous proposal (e.g., lower frame rate or bit rate, fewer bitstream formats or frame resolutions). If node B does not have a counter proposal, then the video negotiations have failed and node B sends node A a message rejecting the conference (step 4710).

If node B does have a counter proposal, then node B sends its counter proposal to node A (step 4712). Steps 4712-4720 are analogous to steps 4702-4710 except that the roles of node A and node B are reversed. The video negotiation processing of FIG. 47 continues until either a mutually acceptable video mode is selected (i.e., successful video negotiation) or nodes A and B are unable to identify a mutually acceptable video mode (i.e., unsuccessful video negotiation).

In an embodiment in which the allowable video modes are contained in a table in the video subsystem, it will be understood that the proposals and counter proposals are constrained to being based on the specific modes in the table.

In theory, the processing within a node for determining whether a proposed unique mode is acceptable and for generating a counter proposal to a particular proposal may depend on one or more of the following influences:

The particular transport(s) that are available (e.g., LAN or ISDN).

The CPU processing bandwidth available for video processing.

The type of hardware installed in conferencing system 100 (e.g., the type of video board 204 of FIG. 2 may influence which bitstream formats are supported).

To a certain extent, these considerations are determined off line when the video subsystem table of mode choices is generated.

Video Manager Negotiation Support

Video manager 516 supports the following three interfaces to support video negotiation.

VSystemAttrMap

DWORD VSystemAttrMap (DWORD, DWORD);

The VSystemAttrMap function takes a DWORD bit rate parameter and a DWORD flags parameter. The bit rate parameter is used by the caller to indicate the maximum communications bandwidth that can be used by a video stream (unidirectional). Units are in kbits per second. The flags parameter is for future use, and is undefined. VSystemAttrMap returns DWORD bitmap used to code a range of video attributes. The returned attributes bitmap is a function of the bit rate specified. For example, the attributes map may differ if the communications media is ISDN or LAN, where the former may allocate approximately 90 kbits for video and the latter may allocate 150-300 kbits for video.

VDecodeAttrMap

VSTATUS VDecodeAttrMap (DWORD, LPVINFO);

The VDecodeAttrMap function decodes bitmap attributes into existing VINFO structure. VDecodeAttrMap takes DWORD bitmap which defines video attributes and a VINFO structure. VDecodeAttrMap modifies (or builds) VINFO structure to reflect the DWORD bitmap defined attributes. The VINFO fields wCaptureFormat, wMaxFrameRate, wWidth, and wHeight may be modified as a result of VDecodeAttrMap. The VINFO structure is of type LOCAL.sub.-- STREAM. The decode produces a unique VINFO structure, and if the input attributes map defines multiple options, then the best case (latest algorithm, highest frame rate, greatest capture resolution) is chosen.

VNegotiate

DWORD VNegotiate (DWORD, DWORD);

The VNegotiate function carries out a negotiation/comparison between two system attribute maps. VNegotiate takes a DWORD bitmap of a local system and a DWORD bitmap of a remote system. VNegotiate returns a DWORD bitmap that is the result of an internal video manager negotiation of video attributes. If the returned value is zero, then negotiation failed. If the returned value is non-zero, then negotiation was successful and the returned value defines attributes common to the two systems. In order that repeated video negotiation is not required, if a non-zero value is returned, then it represents a unique set of video attributes which can be supported at both endpoints. This interface frees upper layer software (e.g., VCI 506) from interpreting and negotiating defined attribute bitmaps.

The following represents and example of the usage of these video negotiation functions:

Participant Information Exchange

The description for this section is the same as the description for the section of the same name in U.S. Patent Application filed Nov. 15, 1994 using U.S. Express Mail Label No. EG029471669.

Caller Attributes and Call Progress

The MakeConnection message contains 64 bytes of user-interpreted data (i.e., a 64-byte field that the user is free to define). A user may use some or all of that field to identify certain attributes of the caller when the caller sends the MakeConnection message to a remote node (callee). These attributes may include the name of the user. In this case, the callee will know who is trying to place a conference call before the conference call is actually accepted by the callee. When the callee is not operating in an auto-answer mode, the callee displays the user-interpreted information contained in the MakeConnection message as part of a dialog box that gives the user of the remote node the option of accepting or rejecting the conference call based on that information.

When a caller attempts to place a conference call to a callee, the duration from the time that the user starts the process (e.g., selecting a "dial the selected phone number" option in a dialog box) until the time that the attempt to place the conference call succeeds or fails may be anywhere from a few seconds to as long as a minute. The caller user is presented with feedback to indicate the progress of the conference call during that duration. Similarly, the callee user is also presented with appropriate call-progress feedback.

The call-progress information presented to the caller user may include audible ringing or a display of simulated ringing (e.g., displaying a graphics image of a phone surrounded by vibration lines). When the attempt to place the conference call is successful, the caller notifies the user with an appropriate graphical display. Since a callee provides a reason code when rejecting a conference call, in that case, the caller displays the reason why the call was rejected to the user.

For the callee, the call-progress information may include a display notifying the callee that a connection has been requested. This display includes the name of the caller user when that information is provided in the user-interpreted data field. The callee also presents a graphical display to the user when the attempt to place the conference call is either successful or unsuccessful, in which case, the caller's reason for rejecting is presented.

Referring now to FIG. 48, there is shown a flow diagram of the call-progress processing when the placement of a conference call is successful. The call-progress processing of FIG. 48 may be summarized as follows:

The caller user uses a dialog box to ask for a conference call to be placed to the callee (step 4802).

The caller network stack puts a connection request packet on the network to the callee (step 4804). It will be understood that steps 4802 and 4804 include all the necessary inter-layer communications from the caller conferencing application through to the caller network stack as shown in FIGS. 5 and 17.

The network stack informs the caller that the connection request was placed on the network (step 4806).

The caller presents ringing to the caller user (step 4808).

Some time after step 4804, the callee receives the connection request packet over the network (step 4810).

The callee sends an acknowledgment packet for the receipt of the connection request packet over the network to the caller (step 4812).

Some time after step 4812, the caller receives the acknowledgment packet from the callee over the network (step 4814).

After step 4812, the callee informs the callee user that the caller is trying to place a conference call, where the callee identifies the caller user to the callee user (step 4816).

The callee user selects the dialog box option to accept the conference call (step 4818).

The callee sends a connection accepted packet over the network to the caller (step 4820).

The caller receives the connection accepted packet over the network from the callee (step 4822).

The caller informs the caller user that the connection has been established (step 4824).

After step 4822, the caller sends a connection accepted packet back to the callee over the network (step 4826).

The callee receives the connection accepted packet over the network from the caller (step 4828).

The callee informs the callee user that the connection has been established (step 4830).

As shown in FIG. 48, the caller presents ringing to the user (step 4808) after it receives acknowledgment from the network stack that the connection request packet has been placed on the network (step 4806). In an alternative embodiment, the caller waits until after receiving the packet acknowledging receipt of the connection request packet by the callee (step 4814) before presenting ringing to the caller user.

If the callee rejects the conference call from the caller (instead of step 4818), then the callee sends a connection rejected packet to the caller over the network (instead of step 4820). In this case, after the caller receives the connection rejected packet (instead of step 4822), the caller informs the caller user that the conference call was rejected and present the callee's reason for the rejection (instead of step 4824). In this case, steps 4826, 4828, and 4830 are omitted.

Conference Participation Messages

The description for this section is the same as the description for the section of the same name in U.S. patent application Ser. No. 08/340,172, filed Nov. 15, 1994, now abandoned.

Flow Control Over Reliable Channels

Referring now to FIG. 36, there is shown a flow diagram of the processing of conferencing systems A and B of FIG. 1 to control the flow of signals over reliable channels. The description for this section is the same as the description for the section of the same name in U.S. patent application Ser. No. 08/340,172, filed Nov. 15, 1994, now abandoned.

Preemptive Priority-Based Transmission

Referring now to FIG. 37, there is shown a flow diagram of the preemptive priority-based transmission processing implemented by the communications subsystem of conferencing system 100 of FIG. 1. The description for this section is the same as the description for the section of the same name in U.S. patent application Ser. No. 08/340,172, filed Nov. 15, 1994, now abandoned.

Rate Negotiation

Referring now to FIGS. 38-41, there are shown state diagrams for the rate negotiation processing. The description for this section is the same as the description for the section of the same name in U.S. patent application Ser. No. 08/340,172, filed Nov. 15, 1994, now abandoned.

Dial Lists

The conferencing application presents to the user a dialog box containing an alphabetized directory of possible callees for conferencing sessions. The user is able to initiate a conference call to one or more callees by selecting from the listed callees.

The conferencing systems support the ability to maintain and access different lists of possible callees. For example, there is a network list maintained by a network administrator and a personal list for each conferencing node. The network and personal lists may overlap (i.e., share a subset of possible callees). The node user is able to access and edit (i.e., read/write/delete) her own personal list, but can only access (i.e., read only) the network list and the personal lists of other nodes.

A node user is presented with the option of (1) accessing her personal list only or (2) accessing a combination (i.e., union) of her personal list and one other selected list (i.e., either the network list or the personal list of another node). When the combined list is selected, it is displayed as if it were a single alphabetized list with the callees from the user's personal list identified in some fashion (e.g., displayed in a different color)

Interrupt-Time Processing for Receiving Data Signals

As described earlier in this specification, video conferencing system 100 of FIG. 5 is implemented under a Microsoft.RTM. Windows.TM. operating system running on host processor 202 of conferencing system 100. Host processor 202 of conferencing system 100 receives data signals from and transmits data signals to hardware components that are external to host processor 202. For example, during a LAN-based audio/video conference, host processor 202 receives compressed video signals from video board 204 and compressed audio signals from audio/comm (ISDN) board 206, and transmits these compressed audio and video signals to LAN board 210 for transmission over the LAN to a remote conferencing system. Similarly, LAN board 210 transmits to host processor 202 the compressed audio and video signals which it receives over the LAN from the remote conferencing system. Host processor 202 then transmits the compressed audio signals to audio/comm (ISDN) board 206 for decompression and playback. The compressed video signals are decompressed by host processor 202.

Since host processor 202 is a serial processor, the various software components that run on host processor 202 operate serially, that is, only one software component operates at a time. The operating system of host processor 202 controls the operations of the various software components by establishing a task schedule by which the operating system allocates the processing time of host processor 202 to the software components.

In addition to being a serial processor, host processor 202 is also a preemptive processor. As a preemptive processor, the operating system is able to suspend the implementation of one function by host processor 202 to have host processor 202 perform a second function. After the second function has completed, the operating system causes host processor 202 to resume the implementation of the first function from where it left off.

An external hardware component (e.g., video board 204, LAN board 210, or audio/comm (ISDN) board 206) can ask the operating system to have host processor 202 perform a particular function by sending host processor 202 an interrupt signal. The operating system may cause the requested function to be performed as a scheduled task of the normal task schedule. In this case, the performance of the task is delayed until the operating system schedules the task. Alternatively, the operating system may perform the requested function preemptively. That is, the operating system may cause host processor 202 to suspend what it was doing long enough for host processor 202 to perform the requested function. In this case, the preemptive processing is said to be implemented during interrupt time.

For example, when LAN board 210 receives compressed audio and video signals over the LAN from a remote conferencing system, LAN board sends host processor 202 an interrupt signal to inform host processor 202 that LAN board 210 has received data signals for host processor 202 to process. In order to provide high quality playback during audio/video conferencing (especially for the audio signals), it is important to process the received data signals as quickly as possible. If host processor 202 were to process the received data signals during a normally scheduled task of the operating system, then the resulting quality of the audio/video conference may be relatively low, since there is no guarantee of a scheduled task executing within a reasonable period of time under the Microsoft.RTM. Windows.TM. operating system.

The audio/video conferencing system processes the received audio and video signals preemptively during interrupt time. By processing the received data signals at interrupt time, the audio/video conferencing system is able to provide higher quality audio and video playback than would otherwise be provided if such processing were implemented as a normally scheduled operating system task.

In a possible embodiment of an audio/video conferencing system, the processing of the received audio and video signals could be attempted to be implemented during the "receive" interrupt time that follows the interrupt signal from LAN card 210 which informs host processor 202 that LAN card 210 has received compressed audio and video signals for host processor 202 to process. It has been discovered, however, that, under Microsoft.RTM. Windows.TM. operating systems, when the audio/video conferencing system attempts to process completely the received data signals during the receive interrupt time, there may be undesirable results. For example, the operating system may crash (i.e., cease proper operations).

Referring now to FIG. 49, there is shown a representation of the interrupt-time processing for receiving data signals by audio/video conferencing system 100 of FIG. 5. When LAN board 210 sends host processor 202 a receive interrupt signal (i.e., informing operating system 4902 that LAN board 210 has received compressed audio/video signals), operating system 4902 suspends the current processing of host processor 202 and passes the interrupt signal to LAN comm stack 560. LAN comm stack 560 passes the interrupt signal to comm manager 518. Comm manager 518 causes the LAN comm software to read the received data signals from LAN board 210. The LAN comm software passes the received data signals to comm manager 518, which stores the received data signals into a queue in memory for subsequent processing. All of this processing occurs during the receive interrupt time that follows the receive interrupt signal. The interrupt processing then terminates and the operating system causes host processor 202 to resume the processing that had previously been suspended.

Operating system 4902 receives regular clock interrupt signals (e.g., every approximately 10 milliseconds). Operating system 4902 passes these clock interrupt signals to comm manager 518. Comm manager 518 uses the clock interrupt time (which follows a clock interrupt signal) to complete the processing of any received audio and video signals that were queued into memory during the previous receive interrupt time. This clock-interrupt-time processing includes passing the received audio and video signals from the comm manager 518 to the transport independent interface 510, which distributes the audio signals to the audio manager 520 and the video signals to the video manager 516 for decompression and playback processing. After the audio and video signals have been processed, the interrupt processing terminates and the operating system 4902 causes host processor 202 to resume the processing that had previously been suspended.

It has been discovered that this strategy of completing the processing of the received data signals during the clock interrupt time avoids the problems that may be associated with attempting to process completely the received data signals during the receive interrupt time.

Interrupt-Time Processing for Transmitting Data Signals

During a conferencing session, conferencing system 100 generates audio and video signals for transmission to a remote conferencing system. These data signals become available for transmission in discrete packets at discrete moments. That is, the data signals are not generated by conferencing system 100 in a perfectly steady stream. For example, when video signals are generated at a rate of 10 frames per second, a set of compressed video signals corresponding to a video frame is generated once every 100 milliseconds. The set becomes available for transmission only after the entire frame has been compressed. Thus, video signals become ready for transmission in discrete sets at discrete moments.

For typical video frames, each set of compressed video signals is broken up into a plurality of data packets for transmission, for example, over the LAN. If the conferencing system 100 attempted to transmit all of the data packets for a video frame as soon as the data packets were ready, the LAN board would transmit the packets one right after the other with very little time between packets and the remote conferencing system may not be able to receive and process all of the data packets in such a short period of time. As a result, some of the data packets may be dropped by the remote conferencing system causing the quality of the conferencing session to be adversely affected.

Referring now to FIG. 50, there is shown a representation of the interrupt-time processing for transmitting data signals by audio/video conferencing system 100 of FIG. 5. Rather than transmitting all of the data packets as soon as they become available for transmission, conferencing system 100 uses the 10-msec clock interrupt signals (described in the previous section of this specification) to spread out the packet transmission.

As shown in FIG. 50, video board 204 sends a "data-ready" interrupt signal to the host processor 202 to inform the operating system 4902 that a set of compressed video signals corresponding to a video frame has been generated. The operating system 4902 passes this interrupt signal to the video manager 516. During the data-ready interrupt time, the video manager 516 reads the compressed video signals from the video board 204 and passes the compressed video signals to the transport independent interface 510. TII 510 passes the compressed video signals to the comm manager 518, which divides the video signals into packets for transmission over the LAN. Analogous processing is implemented for the compressed audio signals generated by the audio/comm (ISDN) board 206.

No packets are sent over the link to the remote conferencing system during the data-ready interrupt time. Rather, packets are sent during the interrupt times that follow the 10-msec clock interrupt signals. When the comm manager 518 receives a clock interrupt signal from the operating system 4902, if there are one or more packets ready to be sent, then the comm manager 518 sends a single packet to the LAN comm software 560 for transmission over the LAN to the remote conferencing system. The result is that the transmission of packets is spread over time with at most one packet being transmitted every 10 msec.

By spreading the transmission of packets over time, the conferencing system 100 increases the probability that the remote conferencing system will have time to receive and process all of the packets. The result is improved quality of the conferencing session due to the reduced number of packet drops by the remote conferencing system.

Those skilled in the art will understand that other embodiments fall within the scope of the present invention. For example, the time period between clock interrupt signals may be other than 10 msec. In addition, the maximum number of packets transmitted during any one clock interrupt time may be other than 1.

Auto Registration and Remote Confidence Testing

Auto registration is a process, along with appropriate mechanisms and utilities, for electronically registering a video conferencing product. The registration information is deposited in a customer support database for registered products. Auto registration is coupled with remote confidence testing, through which customers can test their ISDN line as well as the basic functionality of their conferencing product.

A purpose of auto registration is to collect registration data for the product manufacturer. Auto registration is designed to increase the percentage of customers who register their products by providing an easy and quick user interaction. The auto registration design allows for future changes and expansions.

Referring now to FIG. 51, there is shown a representation of auto registration environment 5100 for video conferencing. The auto registration environment 5100 comprises a new video conferencing node 100 connected via network 110 to one more confidence test servers 5104. The confidence test servers 5104 are in turn connected via local area network (LAN) 5106 to a network comprising an auto registration database (ARD) node 5110 and a customer support node 5108. It is assumed that, when network 110 is an ISDN network, the ISDN line is connected to new node 100 and has been tested for functionality by the appropriate regional phone company.

New node 100 is operated by a customer who has purchased a video conferencing product according to the present invention. New node 100 comprises the following components/software packages:

A personal computer with an Intel.RTM. i486.TM./33 MHz processor or higher, VGA monitor or better, and at least 8 megabytes of random access memory (RAM);

An audio/comm (ISDN) board such as audio/comm (ISDN) board 206 of FIG. 2;

A LAN board such as LAN board 210 of FIG. 2;

A video board such as video board 204 of FIG. 2;

Video conferencing system software and video conferencing application software, such as those shown in FIG. 5; and

Installation and diagnostics software.

Each confidence test server (CTS) 5104 is a high-speed personal computer with an Intel.RTM. i486.TM. or better processor. The confidence test servers 5104 provide three types of services:

(1) Accepts a registration form from new node 100 and deposits it into the auto registration database of ARD node 5110.

(2) Plays an audio/video segment from a PC/VCR of CTS 5104 to test the basic video conferencing functionality of new node 100. In this case, CTS 5104 behaves exactly like a remote node to the new node. That is, the video and audio streams go through the real-time video conferencing mechanisms that two live nodes would go through.

(3) Downloads an applet (e.g., answering machine software) to new node 100 and installs it in the appropriate directory.

A typical configuration for confidence test server 5104 is as follows:

A personal computer with an Intel.RTM. i486.TM./33 MHz processor or higher, VGA monitor or better, at least 8 megabytes of random access memory (RAM), at least 380 megabytes of hard drive;

An audio/comm (ISDN) board such as audio/comm (ISDN) board 206 of FIG. 2;

A LAN board such as LAN board 210 of FIG. 2;

A video board such as video board 204 of FIG. 2;

A high-speed FAX/Modem 400E (e.g., to dial via a modem into the auto registration database of ARD node 5110);

A 32-bit LAN adapter card;

A personal computer video cassette recorder (PC/VCR) connected to the host processor through the serial port;

Video conferencing system software, such as that shown in FIG. 5; and

Microsoft.RTM. Windows for Workgroup.TM. or other appropriate software.

In an auto registration environment having more than one new node 100, all of the new nodes access the confidence test servers 5104 using the same telephone number. In such an environment, when two or more new nodes 100 are simultaneously calling into two or more confidence test servers 5104, hub 5102 dispatches each incoming call to the next available CTS 5104.

A PC/VCR is connected to the communications port of the host processor of each CTS 5104 via a serial line. The PC/VCR plays an audio/video segment (up to approximately 30 seconds long) to the new node. The audio and video signals from the PC/VCR are input to the audio/comm (ISDN) board and the video board, respectively. In an alternative embodiment, a programmable laser disk is used instead of a PC/VCR. In another alternative embodiment, the confidence test server 5104 is able to record the user of the new node in real time and play back the session to the user as part of the confidence testing. In yet another alternative embodiment, a live operator operates the confidence test server 5104 to provide confidence testing services to the new nodes.

In another embodiment, the CTS consists of a PC that plays an audio-video clip recorded in an AVI (audio-video-interleaved) file. The AVI file is played on a PC running Microsoft.RTM. Video for Windows.TM. software. A video capture board is used on this machine to generate NTSC (National Television Standards Committee) video signals. These video signals are then fed as camera input to a CTS.

Auto registration database (ARD) node 5110 provides the auto registration database into which the confidence test servers 5104 deposit the completed registration forms. The confidence test servers 5104 are connected to ARD node 5110 over a local area network, such as LAN 5106, and the registration forms are transmitted from CTS 5104 to ARD node 5110 over the LAN. Alternatively, a modem may be used for the registration form transfer. The registration forms are deposited into the auto registration database of ARD node 5110 as a transaction to assure data integrity.

Referring now to FIG. 52, there is shown a representation of the architecture for auto registration and remote confidence testing for the new node of FIG. 51. Auto registration software consists of a library module that manages the registration user interface for all products and independent communication modules that support different communication media such as modem and ISDN. The user interface uses the services of the communication modules in a media-independent fashion through a common communication interface. All the necessary information required to create dialog boxes is stored in two files: the ICOMM.INI file and a product-specific table.

As a step in initialization during setup, ICOMM.INI is created. The registration status for each product may be one of two states: SETUP and REGISTERED. On initialization, the status for each product is set to SETUP. The state SETUP is meant mainly for the case where the system is rebooted. Every time the product application is run, it will read ICOMM.INI to check the registration status. In the case of the first run after a reboot, the registration status will have the entry set to SETUP. This will tell the application that it has to bring up the dialog box. Once the dialog box is invoked, the state SETUP has no more meaning. The complete product status entry will be deleted in both reboot and no reboot cases. Thereon, depending on the choice of the user to register or not, the registration status will be modified.

If the user decides to register and if the registration is successful, then the product status entry is created and its value is set to REGISTERED. If registration fails, then there is no action and the product status entry will not be present. In that case, the next time a standalone auto registration program is run, it will fail to find the product status entry in ICOMM.INI and thereby know that the product was not registered earlier.

Auto registration will be invoked at different times depending on whether installation requires the system to be rebooted or not. If the product requires no additional hardware for installation, then auto registration is called as the last step in the setup process. The user's choice to register or not is recorded in ICOMM.INI. When the main application is started again, ICOMM.INI will be read and if the product status entry is SETUP, the auto registration will be invoked again. The user's new choice will be recorded in ICOMM.INI and registration will proceed if required.

If the system needs to be rebooted after installing any hardware, then auto registration will be called only by the main application and not by setup. The application will read ICOMM.INI and find the product status entry to be set to SETUP. In that case, the application will prompt the user to register. The user's new choice will be recorded in ICOMM.INI and registration will proceed if required.

Referring now to FIG. 53, there is shown a flow diagram of the processing for the auto registration and remote confidence testing of auto registration environment 5100 of FIG. 51. This processing is implemented as a client/server application, where the client is a new node 100 and the server is a confidence test server 5104.

New node 100 places a call to connect to a CTS 5104 (step 5302 of FIG. 53) and transmits a registration record (step 5304). In response, CTS 5104 runs the remote confidence test (step 5306). Depending upon the results of the remote confidence test, new node 100 sends CTS 5104 an appropriate acknowledgement/failure message (step 5308). CTS 5104 may then download a free applet onto new node 100 (step 5310), which installs the applet (step 5312). A purpose of the free applet is to encourage new users to register their nodes. After CTS 5104 disconnects (step 5314), CTS 5104 deposits the registration record into the auto registration database of ARD node 5110 (step 5316) and reports any failures to customer support node 5108 (step 5318).

The processing shown in FIG. 53 may be altered to cover a number of possible alternatives for auto registration, such as:

Auto registration only (i.e., no incentives);

Auto registration+confidence testing;

Auto registration+free applet; and

Auto registration+confidence testing+free applet.

Referring now to FIG. 54, there is shown a flow diagram of the processing implemented by the client (i.e., new node 100) for the auto registration processing of FIG. 53. As one of the last steps in installation (or re-installation) and diagnostics for new node 100, a dialog box is displayed to inform the user that he/she may complete the installation by implementing remote confidence testing of the product and registering the product at the same time (step 5402 of FIG. 54). Alternatively, auto registration may be invoked by the main application after the point where the detection of the presence of hardware and drivers is found to be successful. If the user does not want to register the new node 100 and selects the "Cancel" option (step 5404), then the product registration status variable Auto.sub.-- Reg.sub.-- State is set to the value NOT.sub.-- REGISTERED (step 5406) and the auto registration processing is terminated.

If the user does want to register the new node 100 and the "OK" option is selected, then a registration record is prepared (step 5408). A registration record contains user information, extended user information, communications settings, communication port settings, the phone/ISDN number, the product serial number, username/password for server, and the date that the registration record is prepared. The information is stored in the ICOMM.INI file.

User information includes the name, address, phone number, facsimile number, and position of the user. User information is common for the registration of all products of that user. Extended user information, on the other hand, is information that may vary from product to product and includes answers to specific questions posed to the user about the particular product. Extended user information is stored in a separate table for each product.

Communications settings information is the source of information required to set up communications. In the case of modem, it includes port, baud rate, word length, start bits, stop bits, and parity. Communications port settings provides auto registration with information as to how the communications ports are mapped to the PC interrupt request levels. The phone/ISDN number is the server phone number or ISDN number that is to be dialed in order to transmit the registration record.

The product serial number may be stored either on the installation diskettes or in a word in the EEPROM. During setup, the setup program obtains the serial number from the appropriate source and writes it under a SERIAL.sub.-- NUMBER key in the ICOMM.INI file. Since the serial number in ICOMM.INI may get corrupted, if the user decides to register the product at a time other than during the initial setup, instead of reading the serial number form the ICOMM.INI file, the application is requested to read the serial number form the appropriate source and write it to ICOMM.INI. In other words, the serial number is buffered in ICOMM.INI for auto registration to read.

After the registration record is prepared, the user initiates a video conference call from the new node 100 to confidence test server 5104, using a designated toll-free number (step 5408). If the call is not completed (either because the line is busy or because of a line problem (step 5410), then the product registration status variable Auto.sub.-- Reg.sub.-- State is set to the value NOT.sub.-- REGISTERED (step 5406) and the auto registration processing is terminated.

If the connection to CTS 5104 is established, then the new node 100 sends the registration record to CTS 5104, sets the product registration status variable Auto.sub.-- Reg.sub.-- State to the value IS.sub.-- REGISTERED, sets the confidence test timer, and waits for remote confidence testing (step 5412). If the confidence timer times out without receiving any messages or active streams from CTS 5104 (step 5414), then a failure message is displayed (step 5416) and the auto registration processing continues to step 5420.

Otherwise, the new node 100 receives and renders an audio/video clip from CTS 5104 (step 5418). Dialog boxes are then displayed for the user to report to CTS 5104 the audio and video status of the clip and to save the confidence test results (step 5420). The new node 100 then receives and installs the free applet from the CTS 5104 (step 5422). The new node terminates the auto registration processing by disconnecting from the CTS 5104 (step 5424).

Referring now to FIG. 55, there is shown a flow diagram of the processing implemented by a confidence test server 5104 for the auto registration processing of FIG. 53. The CTS 5104 answers the call from the new node 100 and stores the caller's ISDN number (step 5502 of FIG. 55). The CTS 5104 accepts a registration record from the new node 100 and sends an acknowledgement message to the new node 100 (step 5504). The CTS 5104 starts the confidence testing process by sending a message to the new node 100 to display the proper instructions to the user (e.g., "Make sure your head set is connected.") (step 5506). The CTS 5104 then transmits the audio/video clip from its PC/VCR to the new node 100 (step 5508).

After the audio/video clip is over, the CTS 5104 a message instructing the new node 100 to question the user for the audio and video status of the clip (step 5510). If the messages returned from the new node 100 indicate there was a failure in either the audio or video streams (step 5512), then the CTS 5104 sets the CTS.sub.-- State variable to the appropriate value (i.e., AUDIO.sub.-- FAILURE, VIDEO.sub.-- FAILURE, or BOTH.sub.-- FAILED), and prepares and sends a message to the customer support node 5108 (step 5514). If there were no failures, then the confidence test was passed and the CTS.sub.-- Status variable is set to the value CTS.sub.-- PASSED (step 5516). The free applet is then downloaded to the new node 100 and a message is sent to instruct the new node 100 to hang up (step 5518). After the line is disconnected (step 5520), the CTS 5104 prepares an auto registration database transaction and sends the ARD transaction to ARD node 5110 (step 5522). The ARD transaction comprises the registration record, the new node's ISDN number, and the confidence test results.

The unique product serial number is stored in a word in the EEPROM of the product hardware. Alternatively, the product serial number may be stored in the installation disks that are shipped with the video conferencing product. In the latter case, the number is then saved in a location in the EEPROM on the audio/comm (ISDN) board.

The product auto registration status variable Auto.sub.-- Reg.sub.-- State has one of two values: NOT.sub.-- REGISTERED and IS.sub.-- REGISTERED. These values are coded in a word in the EEPROM on the audio/comm (ISDN) board.

The CTS status variable CTS.sub.-- State stores the results of the confidence testing and has one of four values: AUDIO.sub.-- FAILURE, VIDEO.sub.-- FAILURE, BOTH.sub.-- FAILURE, and CTS.sub.-- PASSED. The values are coded in a word in the EEPROM on the audio/comm (ISDN) board.

The auto registration and remote confidence testing procedures may be repeated periodically to register and/or validate system upgrades or updates.

Referring now to FIG. 56, there is shown a representation of the auto registration file format. The auto registration module creates the file "reg.tmp" in the Microsoft Windows.TM. directory to store user data collected from the main auto registration dialog. The reg.tmp file is a temporary file and is deleted when the user exits the auto registration program. The reg.tmp file is transferred from the new node 100 of FIG. 51 to a confidence test server 5104 via modem using X-modem protocol when successful connection is made.

The reg.tmp file comprises a file header followed by blocks of registration data for every product that is being registered in a particular file transfer. The registration data comprises user information and survey data. If there is no survey dialog for a particular product or if the user does not make any selections in the optional survey, then that block of optional data will not be part of the registration file.

In alternative embodiments of auto registration, automatic software upgrades are provided through the infrastructure of the auto registration and remote confidence testing services described in this section. In other alternative embodiments, use of the ISDN board is disabled until the video conferencing product has been registered using the auto registration described in this section. The fields of the reg.tmp file represented in FIG. 56 are described in Appendix K of this specification. Appendix K also describes the data structures for auto registration.

Alternative Embodiments

The description for this section is the same as the description for the section of the same name in U.S. patent application Ser. No. 08/157,694.

In the embodiment of FIG. 2, conferencing system 100 comprises three separate boards: video board 204, audio/comm (ISDN) board 206, and LAN board 210. Those skilled in the art will understand that alternative embodiments of the present invention may comprise other board configurations. For example, an alternative embodiment may comprises a combined video/audio board (for performing video capture and audio capture and playback), an ISDN board (for transmitting and receiving data over an ISDN network), and a LAN board (for transmitting and receiving data over a LAN network).

It will be further understood that various changes in the details, materials, and arrangements of the parts which have been described and illustrated in order to explain the nature of this invention may be made by those skilled in the art without departing from the principle and scope of the invention as expressed in the following claims. ##SPC1##

Claims

What is claimed is:

1. A computer-implemented process for video conferencing, comprising the steps of: (a) negotiating video capabilities by a first node with a second node, wherein the first node transmits a first video capabilities signal comprising two or more fields each field comprising one or more bits and each field corresponding to a different set of video capabilities for the first node; (b) selecting one set of video capabilities for the first node for video conferencing between the first and second nodes; and (c) performing video conferencing by the first node with the second node in accordance with the selected set of video capabilities.

2. The process of claim 1, wherein step (a) comprises the steps of: (1) generating the first video capabilities signal by the first node based on a first-node table that identifies one or more sets of video capabilities for the first node; (2) transmitting the first video capabilities signal from the first node to the second node, wherein the second node generates a second video capabilities signal in response to the first video capabilities signal and transmits the second video capabilities signal to the first node; (3) receiving the second video capabilities signal by the first node from the second node; (4) comparing the second video capabilities signal with the first video capabilities signal by the first node to determine if video capabilities negotiation is complete; and (5) continuing negotiating video capabilities if negotiation is not complete.

3. The process of claim 2, wherein each valid bit of the first video capabilities signal corresponds to a different set of video capabilities in the first-node table.

4. The process of claim 3, wherein each set of video capabilities in the first-node table identifies a frame resolution and a frame rate.

5. The process of claim 4, wherein each set of video capabilities in the first-node table further identifies a bitstream format and a bit rate.

6. The process of claim 3, wherein the first-node table identifies relative preferences of the sets of video capabilities.

7. The process of claim 2, wherein the first-node table identifies relative preferences of the sets of video capabilities.

8. The process of claim 1, wherein each set of video capabilities identifies a frame resolution and a frame rate.

9. The process of claim 8, wherein each set of video capabilities further identifies a bitstream format and a bit rate.

10. An apparatus for video conferencing, comprising: (a) means for negotiating video capabilities by a first node with a second node, wherein the first node is capable of transmitting a first video capabilities signal comprising two or more fields, each field comprising one or more bits and each field corresponding to a different set of video capabilities for the first node; (b) means for selecting one set of video capabilities for the first node for video conferencing between the first and second nodes; and (c) means for performing video conferencing by the first node with the second node in accordance with the selected set of video capabilities.

11. The apparatus of claim 10, wherein means (a) is capable of: (1) generating the first video capabilities signal by the first node based on a first-node table that identifies one or more sets of video capabilities for the first node; (2) transmitting the first video capabilities signal from the first node to the second node, wherein the second node is capable of generating a second video capabilities signal in response to the first video capabilities signal and transmitting the second video capabilities signal to the first node; (3) receiving the second video capabilities signal by the first node from the second node; (4) comparing the second video capabilities signal with the first video capabilities signal by the first node to determine if video capabilities negotiation is complete; and (5) continuing negotiating video capabilities if negotiation is not complete.

12. The apparatus of claim 11, wherein each valid bit of the first video capabilities signal corresponds to a different set of video capabilities in the first-node table.

13. The apparatus of claim 12, wherein each set of video capabilities in the first-node table identifies a frame resolution and a frame rate.

14. The apparatus of claim 13, wherein each set of video capabilities in the first-node table further identifies a bitstream format and a bit rate.

15. The apparatus of claim 12, wherein the first-node table identifies relative preferences of the sets of video capabilities.

16. The apparatus of claim 11, wherein the first-node table identifies relative preferences of the sets of video capabilities.

17. The apparatus of claim 10, wherein each set of video capabilities identifies a frame resolution and a frame rate.

18. The apparatus of claim 17, wherein each set of video capabilities further identifies a bitstream format and a bit rate.

19. A first video conferencing node, comprising: a conferencing subsystem; a communications (comm) subsystem; and a video subsystem, wherein: the conferencing subsystem is capable of negotiating video capabilities with a second node; the comm subsystem is capable of transmitting a first video capabilities signal comprising two or more fields, each field comprising one or more bits and each field corresponding to a different set of video capabilities for the video subsystem; the conferencing subsystem is capable of selecting one set of video capabilities for the video subsystem for video conferencing between the first and second nodes; and the video subsystem is capable of performing video conferencing with the second node in accordance with the selected set of video capabilities.

20. The apparatus of claim 19, wherein: the conferencing subsystem is capable of generating the first video capabilities signal based on a first-node table that identifies one or more sets of video capabilities for the video subsystem; the comm subsystem is capable of transmitting the first video capabilities signal to the second node, wherein the second node is capable of generating a second video capabilities signal in response to the first video capabilities signal and transmitting the second video capabilities signal to the comm subsystem; the comm subsystem is capable of receiving the second video capabilities signal from the second node; the conferencing subsystem is capable of comparing the second video capabilities signal with the first video capabilities signal to determine if video capabilities negotiation is complete; and the conferencing subsystem is capable of continuing negotiating video capabilities if negotiation is not complete.

21. The apparatus of claim 20, wherein each valid bit of the first video capabilities signal corresponds to a different set of video capabilities in the first-node table.

22. The apparatus of claim 21, wherein each set of video capabilities in the first-node table identifies a frame resolution, a frame rate, a bitstream format, and a bit rate.

23. The apparatus of claim 20, wherein the first-node table identifies relative preferences of the sets of video capabilities.

24. The apparatus of claim 19, wherein each set of video capabilities identifies a frame resolution, a frame rate, a bitstream format, and a bit rate.

Patent Citations (36)

Patent	Date	Inventor
US4475193	1984-10-01	Brown
US4509167	1985-04-01	Bantel et al.
US4796293	1989-01-01	Blinken et al.
US4879742	1989-11-01	Taniguchi et al.
US4888795	1989-12-01	Ando et al.
US4905282	1990-02-01	McGlynn et al.
US4953210	1990-08-01	McGlynn et al.
US4996592	1991-02-01	Yoshida
US4998243	1991-03-01	Kao
US5012509	1991-04-01	Nakamura et al.
US5014267	1991-05-01	Tompkins et al.
US5056136	1991-10-01	Smith
US5072442	1991-12-01	Todd
US5073926	1991-12-01	Suzuki et al.
US5131016	1992-07-01	Broughton et al.
US5157491	1992-10-01	Kassatly
US5231492	1993-07-01	Dangi et al.
US5309562	1994-05-01	Li
US5315633	1994-05-01	Champa
US5319793	1994-06-01	Hancock et al.
US5333133	1994-07-01	Andrews et al.
US5335321	1994-08-01	Harney et al.
US5367629	1994-11-01	Chu et al.
US5373549	1994-12-01	Bales et al.
US5374952	1994-12-01	Flohr
US5375068	1994-12-01	Palmer et al
US5384598	1995-01-01	Rodriguez et al.
US5384772	1995-01-01	Marshall
US5392223	1995-02-01	Caci
US5418624	1995-05-01	Ahmed
US5440405	1995-08-01	Ahmed
US5485282	1996-01-01	Takeda
US5488653	1996-01-01	Dimolitsas et al.
US5504837	1996-04-01	Griffeth et al.
US5526037	1996-06-01	Cortjens et al.
EPX523629	1992-07-01

Non-Patent Literature (11)

"Dial M for Modem", by Hill, Alice et al, PC Sources, v2, n7, p343(8), Jul 1991.
"Equipment Catches up with the H .261 Standar", by Gold, Elliot, Networking Management, v9,n8, p46(3), Jul. 1991.
Eleftheriadis et al., "Algorithms and Performance Evaluation of the Xphone Multimedia Communication System". ACM Multimedia 93-Proceedings First ACM International Conference on Multimedia, Aug. 1 , 1993, Anaheim (US).
Zhang et al., "Video Communication on LANs--Multimedia CSCW Applications ". CCECE/CCGEI '93 .
Ramanathan et al., Integrating Virtual Reality, Tele-Conferencing, and Entertainment into Multimedia Home Computers. IEEE Transactions on Consumer Electronics, vol. 38, No. 2, May 1992.
Nakamura et al. , "Personal Multimedia Teleconferencing Terminal ". IEEE International Conference On Communications ICC 90-Session 211. vol. 1, Apr. 1990, Atlanta (US).
Maeno et al., "Distributed Desktop Conferencing System (Mermaid) Based on Group Communication Architecture". IEICE Transactions, vol. E74, No. 9, Sept. 1991, Tokyo JP.
Tanigawa et al, "Personal Multimedia-Multipoint Teleconference System". IEEE INFOCOM 91-Conference On Computer Communications-Session 9D, vol. 3, Apr. 1991, Bal Harbour (US).
Computer Conferencing: IBM Scientists Demo Prototype of Affordable Computer Conferencing System, Nov. 1992. EDGE, on & about AT&T, V7, n223, p.22.
Chen et al., "A Multimedia Desktop Collaboration System", GLOBECOM 92--EEE Global Telecommunications Conference, vol. 2, Dec. 6, 1992-Dec. 9, 1992. Orlando (US).
Vin et al., "Multimedia Conferencing in the Etherphone Environment". Computer, vol. 24, No. 10, Oct. 1991. Long Beach (US).

Abstract

Patent Drawings (57 sheets)