(en)The audio manager provides an interface between one or more upper-level conferencing drivers of the conferencing system and one or more lower-level audio drivers of the conferencing system to isolate the conferencing drivers from the audio drivers. The audio manager is adapted to perform a plurality of functions called by the conferencing drivers. The audio manager comprises a local audio stream state machine and a remote audio stream state machine.

1.ApplicationNumber: US-88756397-A
1.PublishNumber: US-5913062-A
2.Date Publish: 19990615
3.Inventor: VRVILO; BENJAMIN
SLOSS; REED
TUNG; PETER

4.Inventor Harmonized: VRVILO BENJAMIN(US)
SLOSS REED(US)
TUNG PETER(US)

5.Country: US
6.Claims:
(en)The audio manager provides an interface between one or more upper-level conferencing drivers of the conferencing system and one or more lower-level audio drivers of the conferencing system to isolate the conferencing drivers from the audio drivers. The audio manager is adapted to perform a plurality of functions called by the conferencing drivers. The audio manager comprises a local audio stream state machine and a remote audio stream state machine.

7.Description:
(en)INCORPORATION BY REFERENCE
This application is a continuation of application Ser. No. 08/390,560 filed on Feb. 17, 1995, now abandoned, which is a continuation of U.S. patent ppplication Ser. No. 08/340,173, filed Nov. 15, 1994 U.S. Pat. No. 5,574,934, which is a continuation-in-part of U.S. patent application Ser. No. 08/157,694, filed Nov. 24. 1993 U.S. Pat. No. 5,506,454, which is incorporated herein in its entirety by reference. This application is also related to U.S. patent application Ser. No. 08/305,206 (filed Sep. 13,1994) U.S. Pat. No. 5,600,684, U.S. patent application Ser. No. 08/137,319 (filed Oct. 14, 1993) U.S. Pat. No. 5,452,299, and in U.S. patent application Ser. No. 08/170,146 (filed Dec. 20, 1993) U.S. Pat. No. 5,581,702, which are all incorporated herein in their entireties by reference.


MICROFICHE APPENDIX
Appendices A through E are included in a microfiche appendix submitted as a part of this application. The microfiche appendix has one (1) total microfiche sheets and fifty-two (52) total frames.
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® Windows™ operating system. There are difficulties, however, with providing real-time conferencing in non-real-time windowed environments.
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.
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® Windows™ operating system.
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 an audio manager for a computer-based conferencing system. According to a preferred embodiment, the audio manager provides an interface between one or more upper-level conferencing drivers of the conferencing system and one or more lower-level audio drivers of the conferencing system to isolate the conferencing drivers from the audio drivers. The audio manager is adapted to perform a plurality of functions called by the conferencing drivers. The plurality of functions comprises a first function for retrieving the number of different audio managers installed in the conferencing system; a second function for opening a local or remote audio stream; a third function for controlling capturing of the local audio stream; a fourth function for controlling monitoring of the captured local audio stream; a fifth function for controlling linking between a network and the captured local audio stream; a sixth function for controlling linking between the network and the remote audio stream; a seventh function for controlling playing of the remote audio stream; an eighth function for returning a packet number for a current audio packet of the local or remote audio stream; a ninth function for controlling muting of the local or remote audio stream; a tenth function for adjusting characteristics of the local or remote audio stream; an eleventh function for returning status of the local or remote audio stream; a twelfth function for returning capabilities of the audio manager; a thirteenth function for registering an audio stream monitor; a fourteenth function for closing the local or remote audio stream; and a fifteenth function for shutting down the audio manager and the audio drivers.
The present invention also comprises an audio manager for a computer-based conferencing system. According to a preferred embodiment, the audio manager provides an interface between one or more upper-level conferencing drivers of the conferencing system and one or more lower-level audio drivers of the conferencing system to isolate the conferencing drivers from the audio drivers. The audio manager comprises a local audio stream state machine and a remote audio stream state machine.


BRIEF DESCRIPTION OF THE DRAWINGS
Other objects, features, and advantages of the present invention will become more fully apparent from the following detailed description of the preferred embodiment, 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, according to a preferred embodiment of the present invention;
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 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 a preferred embodiment of the hardware configuration of the audio/comm 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 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;
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 connection setup and teardown sequences;
FIGS. 30 and 31 are diagrams of the architecture of the audio/comm board;
FIG. 32 is a diagram of the audio/comm 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, according to a preferred embodiment of the present invention;
FIG. 39 is a stat e 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.


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 an integrated services digital network (ISDN) 110. 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 ISDN 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. In a preferred embodiment, 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 head set.
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, according to a preferred embodiment of the present invention. Each conferencing system 100 comprises host processor 202, video board 204, audio/comm board 206, 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, according to a preferred embodiment of the present invention. 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 board 206 of FIG. 2, according to a preferred embodiment of the present invention. Audio/comm 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.
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, according to a preferred embodiment of the present invention. 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 board 206 of FIG. 4. All of the other software modules depicted in FIG. 5 reside and run on host processor 202 of FIG. 2.
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. 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, 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.
During conferencing, audio I/O hardware 410 of audio/comm 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. Comm task 540, also running on DSP 406, then formats the compressed audio for ISDN transmission and transmits the compressed ISDN-formatted audio to ISDN interface 402 for transmission to the remote site over ISDN 110.
ISDN interface 402 also receives from ISDN 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, audio capture/compression and decompression/playback are preferably performed entirely within audio/comm board 206 without going through the host processor. As a result, audio is preferably continuously played during a conferencing session regardless of what other applications are running on host processor 202.
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 communications (comm) manager 518 using comm application programming interface (API) 510. Comm manager 518 passes the compressed video through digital signal processing (DSP) interface 528 to ISA bus interface 408 of audio/comm 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 110.
ISDN interface 402 also receives from ISDN 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 using 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® Windows™ operating system for eventual display in a video window on monitor 106.
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. Comm manager 518 passes the 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 110.
ISDN interface 402 also receives from ISDN 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® Windows™ GDI (not shown) for display in a data window on monitor 106.
Preferred Hardware Configuration for Conferencing System
Referring now to FIG. 6, there is shown a block diagram of a preferred embodiment of the hardware configuration of audio/comm board 206 of FIG. 4. Referring now to FIGS. 30 and 31, there are shown diagrams of the architecture of the audio/comm board. Referring now to FIG. 32, there is shown a diagram of the audio/comm board environment. 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.
Software Architecture for Conferencing System
The software architecture of conferencing system 100 shown in FIGS. 2 and 5 has three layers of abstraction. A computer supported collaboration (CSC) infrastructure layer comprises the hardware (i.e., video board 204 and audio/comm board 206) and host/board driver software (i.e., video/host interface 526 and DSP interface 528) 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 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 506, 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 554 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 layer. The DSP interface 528 is hidden in the datalink module (DLM). 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 the transmission priority scheme.
In a preferred embodiment in which conferencing system 100 performs software video decoding, video capture driver 522 is implemented on top of video/host interface 526 (the video driver). In an alternative preferred embodiment in which conferencing system 100 performs hardware video decoding, a video 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.
In a preferred embodiment, conferencing system 100 is implemented mostly in the C++ computer language using the Microsoft® Foundation Classes (MFC) with portions implemented in the C7.0 computer language.
Audio/Video Conferencing Application
Audio/video conferencing application 502 implements the conferencing user interface. Conferencing application 502 is implemented as a Microsoft® Windows™ 3.1 application. One child window will display the local video image and a second child window will display the remote video image. Audio/video conferencing application 502 provides the following services to conferencing system 100:
Manages main message loop.
Performs initialization and registers classes.
Handles menus.
Processes toolbar messages.
Handles preferences.
Handles speed dial setup and selections.
Connects and hang up.
Handles handset window
Handles remote video.
Handles remote video window.
Handles local video.
Handles local video window.
Handles call notification dialog box.
Plays sounds.
Handles address book lookup for caller identification purposes.
Audio/video conferencing application 502 is made up of six main modules:
(1) ABIF.LIB: An address book class wrapper library.
(2) SUBDLG.LIB: A sub-dialog library.
(3) CMDLL.DLL: A conference manager DLL (shown in FIG. 5 as conference manager 544).
(4) CMIF.LIB: A library with high-level interfaces to CMDLL.DLL and other sub-systems.
(5) PSVIDEO.EXE: An audio/video application proper.
(6) PSNOTIFY.EXE: An incoming call notification application (a listener).
CMIF.LIB
The CMIF.LIB CMIF was created to encapsulate several subsystem APIs: conference manager API 544, conferencing API (VCI) 506, audio API 512, and video API 508. It is a set of (mostly) static classes (data and functions). It consists of 3 classes: CCm, CImageSize, and CImageState. CImageSize encapsulates the various sizes that video images can have. CImageState encapsulates the different attributes and drivers that a video image can have. CCm handles everything not covered by CImageSize and CImageState. CCm includes a callback function to handle notification from CMDLL.
The callback function in CCm can be called in the context of another application (namely VTHREAD.EXE of VCI 506). This is the reason that CMIF is a separate library. By making it a separate library the compile flags are set to use smart callbacks in a way that is compatible with MakeProcInstance(). MakeProcInstance() is used on the callback to force the data segment to point appropriately upon entering the callback function.
CCm
The CCm class simplifies calling routines in subsystem APIs. It has the following capabilities: loading and unloading of the subsystem DLLs, registering and unregistering with CMDLL, call support, channel pair support, stream support, CMDLL callback handler, and NO -- VCI support. CCm assumes that there is one video call. This call can be in various different states: disconnected, connecting, connected, and disconnecting. Tests and assertions are performed throughout the code to verify that operations on a call can be carried out during the current call state.
Loading and Unloading
CCm::Load() loads the various subsystems. It does so by calling LoadLibrary() for each DLL. For each subsystem, after successfully loading the library into memory, CCm::Load() loads a table of function pointers for each subsystem by calling GetProcAddress(). There are macros that allow CCm to call these functions in a manner consistent with implicit linking.
CCm::Unload() unloads the various subsystems. It does so by calling UnloadLibrary() for each DLL.
Registering and Unregistering
CCm::Register() registers CCm::Callback() as a callback function with CMDLL 544. It also registers an application window with CMIF to post messages to. It should be called once upon application startup.
CCm::UnRegister() unregisters with CMDLL. It should be called once upon application shutdown.
CCm::Install() allows an application that is not currently installed in CMDLL's registry to install itself there. This should be called if CCm::Register() fails with a return code of CMRC -- UNKNOWNAPP. Then, after successfully calling CCm::Install(), call CCm::Register() again.
Call Support
CCm::MakeCall() allows the application to place a video call via CMDLL 544. CCm::MakeCall() takes a ISDN phone number (of the callee) and a name (of the caller). The name is passed to the callee for display in the caption of the remote video window. Upon successful return from CCm::MakeCall(), the call will be in the connecting state.
CCm::HangupCall() hangs the current call up. If there is no call in progress, it returns an error code. Upon successful return from CCm::HangupCall(), the call will be in the disconnecting state.
Channel Pair Support
A channel pair is a bidirectional VCI/TII communication channel. It can be used to send and receive data from a peer application during a video call. Channel pairs are established with a peer application (facilitated through registration) and are referred to by an id (0-31). CMDLL 544 provides this channel pair support to CMIF.
CCm::GetChannelPairInfo() can be called with a channel pair id. It will return information pertaining to the given channel pair or an error code if it failed.
CCm::GetChannelPair() requests a channel pair from CMDLL. The request is completed when CMDLL sends the CMN -- CHANNELPAIR callback message. This message indicates whether or not the channel was successfully opened. Once a channel pair is opened it needs to have a channel handler registered for it. This is done by calling CCm::RegisterChanHandler() for each channel in the channel pair.
After successfully registering a channel handler for each channel in a channel pair, the channel pair can be used to send and receive data.
CCm::SendData() requests that data be sent on the outbound channel within a channel pair. The request is complete when VCI sends the CHAN -- DATA -- SENT callback message to the channel handler.
CCm::ReceiveData() posts a receive buffer to VCI/TII for the inbound channel to receive data into. When data has been successfully received into a buffer for the inbound channel, VCI sends the CHAN -- RCV -- COMPLETE callback message to the channel handler. This buffer must not be freed until the receive has completed.
When finished using a channel pair, call CCm::CloseChannelPair() to request that the resources associated with it be freed. The request is complete when CMDLL sends the CMN -- CLOSECHANNELPAIR callback message. After receiving this message do not attempt to use the channels in that channel pair.
PSVIDEO encapsulates registering a channel handler and sending and receiving data in a class called CChannelPair.
Stream Support
VCI stream groups contain both an audio data stream and a video data stream. They can be either local or remote. Local stream groups are created by calling CCm::CreateLocalStream() and remote stream groups are created by calling CCm::CreateRemoteStream(). Both take as parameters a window in which to play the video, a size to play the video at, a state to play the video in, and flags specifying whether or not to play the video/audio. Local stream groups do not play audio locally, whereas remote stream groups do. There is preferably only local stream group and one remote stream group.
To start sending the local stream group to the peer video conferencing application, call CCm::SendLocalStream(). This should be called only once after the video call has been established and the local stream group has been created.
To start and stop sending during a video call, call CCm::SendLocalStreamMute(). This has the same syntax as the former, but does not stop the audio time stamps from being sent to the peer application. These audio time stamps are preferably continually sent, after they have been started for the life of a video call.
To start and stop playing either the local stream group or the remote stream group, call CCm::PlayLocalStream() or CCm::PlayRemoteStream(), respectively. These also can be used to play a stream at a different size or state than it was previously being played at before (i.e., the local stream group can be played as an IRV video stream or a YUV video stream).
To change the window in which either the local stream group or the remote stream group is being played, call CCm::RedirectLocalVideo() or CCm::RedirectRemoteVideo(), respectively.
To take a snapshot of either stream, call CCm::SnapStream(). It will return a device independent bitmap of the given stream group's current video image. It does not matter what the size or state of the video stream is or whether it is the local or remote stream group.
To control the volume of the remote stream group (remember local streams do not play audio), call CCm::SetWaveVolume().
To switch the remote stream group in and out of open audio (half-duplex speaker phone), call CCm::SetOpenAudio().
To control either stream groups video attributes (color, contrast, brightness, and tint), call CCm::SetColor(), CCm::SetContrast(), CCm::SetBrightness(), and CCm::SetTint(). However, CCm::SetTint() is not preferably supported for the remote stream group.
To adjust the display quality (frequency) of the local stream group, call CCm::SetFRC(). Also, local stream groups can toggle between mirror and camera views and normal and wide angle views.
To toggle these attributes, call CCm::SetMirrorView() and CCm::SetWideAngle(), respectively.
CCm::Redraw() forces the given stream group to redraw itself.
To free the resources allocated for a stream group call CCm::DestroyStream( ).
CMDLL Callback
When registering with CMDLL 544, a callback function is specified. CMIF specifies a procedure instance for CCm::CallBack(). This callback function may be called in the context of another application. This is why MakeProcInstance() is used to generate a procedure instance for the callback. However, smart callbacks (e.g., Microsoft® C8.0 compiler flag /GA) are not compatible with MakeProcInstance(). Use the old style callbacks (e.g., Microsoft® C8.0 compiler flag /Gw, /GW or /GA/GEea) instead.
This callback function receives the following types of messages from CMDLL 544: call messages (CMN -- CALL, CMN -- HANGUP, CMN -- REJECTED, CMN -- PROGRESS), channel pair messages (CMN -- CHANNELPAIR, CMN -- CLOSECHANNELPAIR), CMDLL shutdown message (CMN -- EXIT) and error message (CMN -- ERROR). For each message sent to this callback, a message is posted to the application window that was given during registration.
When a call message is received, the call state is tested to validate whether the message is valid in the current call state or not. When a message is received that does not correspond to the current call state, a debug message is printed out and in some cases an appropriate action is taken to correct the situation. When a CMN -- CALL, CMN -- HANGUP, or CMN -- REJECTED message is received, the call state is updated to reflect the state diagram shown above.
NO VCI Support
The NO -- VCI compiler is a user interface (UI) development tool to help UI developers debug and test their code without the subsystems running. Most every function that calls a subsystem has the following code fragment:

______________________________________// code to handle either case#ifindef NO.sub.-- VCI// code to handle the case when calling the subsystem..#else// code to handle the case when not calling the subsystem...#endif// code to handle either case______________________________________

As the subsystems change, this flag allows the UI developers to postpone integration with these changes. Also, debugging is sometimes difficult when running with the subsystems enabled.
Miscellaneous
CCm::GetUtilization() queries the video subsystem for the cost (as a percent of system utilization) of a single video window with the given characteristics. If there is a failure in getting this cost, the cst returned is 100.
CCm::GetErrorText() gets the error text for a given error code from either CMDLL 544 or VCI 506.
CCm::IsLostCallError() returns TRUE if the given error code could represent a lost call.
CImageSize
CImageSize encapsulates the image size and display speed attributes of a video stream. It contains operators that allow for iteration through the various sizes. Lastly, it contains member constants that can be queried to see if various image attributes are supported for a given video device driver at the size represen ted by the object.
CImageState
CImageState encapsulates the image state of a video stream. This state is represented in terms of features that can be turned on or off and various video device drivers (e.g, IRV and YUV). Calling CImageState::AddFlag() will turn on or off a given feature in the current image state. To test to see if a feature is on or off, call CImageState::IsFlagSet(). CImageState also allows the changing of the current video driver.
PSVIDEO EXE
This section provides details on the objects and class hierarchies in PSVIDEO.EXE. The objects are described in a top down fashion. The class structure is defined with the following two long-term goals:
(1) The architecture is extensible to a multipoint environment.
(2) The architecture is easy to modify. For example, it is easy to rip out all video specific code in order to implement a phone-on-the-screen.
Frame, View, and Image
Three terms are used quite a bit in this section: frame, view, and image windows. The image window corresponds to the area of the display containing just the video image. The view image contains the image window plus a panel which has buttons and controls specific to the view. The frame is the top-most window and contains the local view.
Class Descriptions
This section describes most of the classes in the class hierarchy, as well as their relationships to other classes.
CCyApp
CCyApp is derived from the Microsoft® CWinApp class, which is the base class from which a Microsoft® Windows™ application object is derived in the MFC framework. The CCyApp object provides member functions for initializing the application and for running the application. This object is constructed when other C++ global objects are constructed and is already available when the Microsoft® Windows™ operating system calls the WinMain function, which is supplied by the MFC framework. As required by MFC, the Initinstance member function has been overridden to create the application's main window object.
Many classes in PSVIDEO declare a CSettings class. Each class is responsible for reading, updating, and saving their own settings (e.g., window screen location) or preferences (always on top). They do this through the CSettings class. Each class that requires settings to be saved implements DoSettings(CSettings *). The DoSettings(CSettings *) function gets and puts the settings associated with that class. When settings are loaded, changed, or saved, parent classes are responsible for notifying their children about the setting event. The children then call their children until all the classes in the hierarchy have performed the necessary setting adjustments. Use the Visual C++ browser to look at examples of DoSettings implementations.
A preference is a value that is stored in the PSVIDEO.INI file under the Preferences! section. A setting is a value that is stored in the PSVIDEO.INI file. The difference is that a preference is set in the Preference dialog box, while settings are not. Settings are things like window size and position.
The CSettings class is derived both from ISettings and CCyDialog. ISettings is not derived from any thing. It declares the virtual function DoSetting (which is implemented by classes derived from ISettings). CSettings also provides a dialog box for user preferences.
CCyApp is also responsible for handling <F1> and <Shift-F1> help requests. Because PSVIDEO uses the Microsoft® Viewer for displaying help instead of the standard Microsoft® Windows™ help engine (WinHelp), the void CWinApp::WinHelp function is overridden so that the viewer is called instead of WinHelp.
When the application is invoked, the MFC framework calls CWinApp::InitInstance. When the application is closed, the MFC framework calls CWinApp::ExitInstance.
CCyFrameWnd
CCyFrameWnd is derived from CFrameWnd and is not directly instantiated. The CFrameWnd class provides the functionality of a Microsoft® Windows™ single document interface (SDI) overlapped or pop-up frame window, along with members for managing the window. The main reason for defining CCyFrameWnd is to provide a class with methods common to the three windows in the application (handset, local video, and remote video), including the non-client drawing and system menu handling.
Frame windows have access to the video controller. The main reason for this is that the CAppFrame class (derived from CCyFrameWnd) informs the video controller about certain events (e.g., call started, call ended) and queries the video controller for certain information (e.g., whether the video windows are ready to close).
CCyAppFrame
CCyAppFrame is derived from CCyFrameWnd. This class implements the PSVIDEO main window (which includes the handset) and owns all dialog boxes in the application (except the Preferences . . . dialog box in CSettings, which is available through its reference to CSetting). This class is responsible for handling menu selections and handset button clicks.
CMIF contains a callback routine for processing callbacks from CMDLL 544. CMIF passes CMDLL messages on to CAppFrame. CAppFrame implements the following message handlers for the messages passed on from CMIF:
CAppFrame::OnCmCall--A call ha s started. Notify CHandset so that the appropriate call progress is displayed and the call duration clock starts. Notify CVideoController so that it starts to play the remote video stream. The caller also requests a channel pair from CMDLL. This channel pair is used for sending and receiving control data such as mute on or mute off.
CAppFrame::OnCmHangup--A call has ended. Inform CHandset so that call progress is set correctly and the duration timer stops. Notify CVideoController so that it stops playing the remote video stream.
CAppFrame::OnCmRejected--The call that was placed was rejected by the other side. Inform CHandset so that call progress is set correctly.
CAppFrame::OnCmProgress--Call progress messages originating from VCI 506. Inform CHandset so that call progress is set correctly.
CAppFrame::OnCmChannelPair--A channel pair has been established. Register channel handlers for the input and output channels.
CVideoFrame
CVideoFrame encapsulates properties and methods for a video window. It is derived from CCyFrameWnd and is not directly instantiated. Two classes are based on CVideoFrame: CLocalFrame and CRemoteFrame. The CVideoFrame class' main responsibility is to manage and keep track of frame window size and handle commands specific to video frame windows.
CVideoController
The CVideoController object is derived from CCmdTarget and ISettings. Its main purpose is to manage video windows. In a preferred embodiment of the present invention, there are only two video windows. In alternative preferred embodiments, there may be several video windows and this class will then be extended to keep track of the list of available video windows.
The CVideoController class is also responsible for implementing size and display rate restrictions. When the user resizes a video frame window, CVideoController will determine appropriate size for the "other" window based on system characteristics (CPU and graphics adapter). CVideoController is responsible for sending size messages to the appropriate video window. CVideoController keeps track of state information related to audio and video, e.g., mute, high quality, and open audio.
Auto-Sizing of Video Windows
The audio/video conferencing application 502 has two video windows: one for the local camera video image and one for the remote camera video image. The user of application 502 expects to be allowed to change the size of these video windows. However, the computer system may not have enough processing power to accommodate such a change, due to the demands larger video windows place on the system. Specifically, increasing the size of one of the video windows may exceed an acceptable amount of the computer system's processing power. If the user is allowed to change the size of one of the video windows without any constraints, the video quality may degrade to an unacceptable level. Conferencing application 500 automatically resizes the windows to utilize some acceptable amount of the computer system's processing power. The sizes a video window can be are preferably constrained to an optimized set based on the video subsystem software.
Split and Combined Modes
Conferencing application 502 can be in one of two modes: split or combined. This section explains how the frame, view, image model is used to implement split and combine. In general, a method is provided to convert a single top-level window, W, with more than one child window, w1, . . . ,wn, to multiple top-level windows, W1, . . . ,Wn, each of which contains one of the child windows w1, . . . ,wn. Typically, the window W is the same window as some window Wi, 1<=i <=n. A top-level window is a window that is the child of the desktop window. The method assumes that in the single window mode, there is a top-level window with several child windows. Each child may have its own children, but this is irrelevant to this discussion. In split windows mode, each child (or a subset of the children) is re-parented and gets its own parent (which is a top-level window). These additional top-level windows can be managed in at least two ways:
(1) Create them upon application start-up. If the application is in combined window mode, then hide the windows. If the application is in split windows mode, then show the windows.
(2) Dynamically create and destroy them as needed.
Switching between combined window mode and split windows mode may be implemented as follows:
SingleToMultiple:
Create top level windows if needed.
Re-parent children that needs to be re-parented.
Show the additional top-level windows.
MultipleToSingle:
Hide the top level windows that are losing their child windows.
Show the children in the single window.
Destroy top-level windows if needed.
In PSVIDEO.EXE, the split/combine code can be found in the file videoctl.cpp in the method CVideoController::SetGrouping.
Control Channel Management
This section describes the application control channel. The control channel enables conferencing application 502 to inform peer of events (e.g., mute on and mute off) and transfer arbitrary size information. In general, the application control channel can be used to transmit any data. For example, conferencing application 502 has an "audio/video mute" feature, which allows a video conferencing participant to click on a button so that he/she is no longer visible to the other conference participant. The peer application must be informed about this operation so that it can discontinue displaying the video image and instead display some indication that the other side turned mute on.
By establishing a control channel and a control structure, peer applications are able to be notified about events that require some action. The control channel assumes reliable transfer of control packets. Conferencing system 100 does not confirm the receipt of control channel messages.
This simplifies the implementation. All control channel messages are of the following form:

__________________________________________________________________________typedef struct tagChanMsgData {WORD wMsg; // message identifier.DWORD dwMsgParam; // message specific parameter/value.DWORD dwSequence; // sequence number for a series of control packets.WORD wBufferSize; // size of the buffer following the structure.} ChanMsgData, FAR* LPCHANMSGDATA;__________________________________________________________________________

In order to send additional data along with a control channel structure, data is arranged following the structure. The size amount of data is specified, in number of bytes, by the dwBufferSize field. Four messages have been defined: mute, high quality snapshot, application launch, and application launch response.
Mute Message
Only wMsg and dwMsgParam are used for the MUTE message. wMsg: value=1.
dwMsgParam: LOWORD of dwMsgParam specifies the audio state.
0=audio mute off
1=audio mute on
2=no change
HIWORD of dwMsgParam specifies the video state.
0=video mute off
1=video mute on
2=no change
For example, to mute video without changing the state of audio dwMsgParam would be MAKELONG(2, 1 ). CVideoController sets the mute state when the user clicks the mute button and informs the peer about its new state using the control channel.
High-Quality Snapshot Message
The SNAPSHOT message is sent to the peer application when a still image capture that disables outgoing video stream is in progress. For example, to achieve a certain still image quality, outgoing video may have to be disabled. In these situations, the SNAPSHOT message is sent to peer application. The message is resent when the flow of outgoing video resumes. Only wMsg and dwMsgParam are used for the SNAPSHOT message. wMsg: value=2.
dwMsgParam: LOWORD of dwMsgParam specifies the snapshot state.
0=High-quality mode OFF
1=High-quality mode ON
Application Launch
The APPLAUNCH message is sent to a remote node to launch an application. The application path is specified in the buffer following the control channel structure. wBufSize holds the size of the buffer. The other fields are not used. wmsg: value=101.
Application Launch Response
The APPLAUNCHRESPONSE message is sent by the remote node that was asked to launch an application. The return code of the Microsoft® Windows™ WinExec function is passed back in dwMsgParam. The other fields are not used. wMsg: value=102.
CChanPair
CVideoController has an instance of a CControlChannel class which is derived from the class CChanPair. CChanPair provides an abstraction of CMDLL and TII services.
Video View Class Relationships
The CVideoView class' main responsibilities are to manage image (video) windows (CImageWindow) and handle the controlb bar operations available in the video panels (CVldeoViewPanel). There exist two classes, CLocaView and CRemoteView, that encapsulate specific functionality of the remote and local windows. The CLocalView and CRemoteView classes have a CLocalImage and CRemoteImage class, respectively. In addition, CLocalViewPanel and CRemoteViewPanel have been derived to contain functionality specific to the local and remote windows.
CVideoView know s about the CVideoController class. When the user selects a function such as size or snapshot in the panels, the CVideoView informs the CVideoController about the user's request. CVideoController is responsible for auto-sizing the video windows.
Handset Class Relationships
As described earlier, the CAppFrame class has an instance of a CHandset. The CHandset class is a window. The controls inside the handset window are specified in a dialog template. This is the reason why CHandset is derived from CDialog. A modeless and borderless dialog box is placed on top of a window, in this case, CAppFrame. CHandset is responsible for handling all button clicks and user interactions in the handset area.
In addition to using several MFC defined classes, CHandset also has a CLcd and a speed dial list, a CSDComboBox. The CLcd class consists of instances of CStatic for call progress information and call duration, and of CEdit for entering the number to dial. CSDComboBox is an owner-drawn listbox that displays the five most recently dialed numbers and also speed dial numbers. Finally, CHandset contains several buttons.
An interface is defined for interactions between CHandset and CSpeedDial. When a number has been dialed, the CHandset informs CSpeedDial about the dialed number. CSpeedDial is then responsible for updating the speed dial list (the CSDComboBox) and the PSVIDEO.INI file.
Dialog Boxes
Conferencing application 502 contains several dialog boxes. The focus of this section is to describe CCyDialog. A special dialog box class is derived from CDialog to avoid problems when dialog boxes are displayed when the application is in split mode and some or all of the topmost windows (handset, local video, remote video, snapshot) are always-on-top. Specifically:
(1) If a dialog is brought up when the application was in split mode (3 Windows) and one of the topmost windows is always-on-top, then portions of the dialog box would otherwise be obscured by the always-on-top window. For example, the Preferences dialog box would be on top of the handset window but underneath the remote window.
(2) When the user brings up a dialog box when the applications is in split mode (3 Windows), the user would otherwise be able to click on the local window and then obscure the dialog box.
The problem with (1) is that the OK and Cancel buttons would often be hidden by the video windows, so the application would be difficult to use. The problem with (2) is that a user could bring up a modal dialog box but perform functions outside of the dialog box before closing it.
To make the application easier to use and more consistent with Microsoft® Windows™ operating system standards, CCyDialog is introduced and the problematic dialog boxes are derived from CCyDialog instead of CDialog. As a result, problems (1) and (2) are fixed. When a dialog box derived from CCyDialog is initialized, if the ProShare application has any visible topmost windows and the dialog box's owner is not topmost (e.g., the local window is topmost, but the owner of the Preferences is the handset window), then the dialog box owner window is made topmost to ensure the dialog is visible. To prevent the user from clicking in other application windows, all other topmost windows are disabled. These operations are reversed when the dialog box is destroyed.
When the user switches between conferencing application 502 and other applications, special care is taken due to the modifications made in CCyDialog::OnInitDialog. When conferencing application 502 is deactivated, CCyDialog::OnActivateApp turns temporary topmost off, and then back on when the application is re-activated. In addition, CCyDialog::OnActivateApp reenables the user interface while conferencing application 502 is deactivated so the user can click on any of the conferencing application's visible windows to get back to the dialog (and re-activate conferencing application 502).
Helper Classes
Dialog Helper
The dialog helper class, CDialogHelper, provides a method for obtaining dialog boxes with exact pixel dimensions in the Microsoft® Windows™ environment. The Microsoft® Windows™ operating system provides support for dialog boxes, both in the API and in the support tools (dialog editors), but the coordinates used for dialog boxes and their elements are not pixels but "dialog units," which are calculated according to the font and display driver in use. Dialog boxes are not created with exact pixel dimensions and coordinates using this support, yet this is often desired, for example, when the position of dialog elements relates to a fixed size live video display (160×120 pixels).
The CDialogHelper class simplifies the manipulation of dialog boxes. If a dialog box is designed with the assumption that 1 dialog unit=1 pixel, CDialogHelper can use a dialog template to resize and position dynamically the dialog elements correctly. The procedure follows:
Create the dialog box initially invisible.
Load the dialog template used to create the dialog box.
For each control referenced in the template,
Get a handle to the actual control, and
Use the coordinates from the template to reposition and resize the control.
Use the coordinates from the template to resize the entire dialog box.
Unload the dialog template.
Make the dialog box visible.
By using a dialog template that is a subset of the template used to create the dialog box, it is possible to resize and reposition only some of the controls, enabling a combination of pixel and dialog units to be employed.
Fast Bitmap Buttons
An owner-draw button class, CFastBmpBtn, solves the following problems:
(1) Microsoft® Windows™ operating system provides basic button objects in several styles (particularly push buttons, check boxes, and radio buttons) with a default graphical appearance that is not easily customized. It is possible for an application to use "owner-draw" buttons, but then all distinction between button types is lost and the application must implement check box and radio button functionality itself. The Microsoft® Foundation Classes (MFC) provide a bitmapped button class that allows an application to provide up to four images to represent the various states of a button (up, depressed, with focus, and disabled), but this does not solve the basic problem, and each image must be stored in a separate bitmap, making maintenance difficult.
(2) When a user clicks on a Microsoft® Windows™ button with the mouse cursor, the button takes the input focus even if the user moves the cursor off the button before releasing the mouse button, thereby not generating a button press event. This makes it difficult for an application to keep the input focus in a desired location (e.g., an edit control).
The two problems above are solved in the following manner:
(1) CFastBmpBtn, a C++ class derived from the basic MFC window class, allows the developer to start with generic buttons, specifying the styles as desired, then add only three lines of code and one bitmap per button to obtain a flexible graphical appearance. The class dynamically subclasses the existing button object, taking over all standard button actions, in particular the drawing of the button. The bitmap contains 4, 8 or 12 images arranged vertically in a strict order, each representing a different button state (up, depressed, with focus, and disabled) for each of the possible check states unchecked, checked, and indeterminate check. The appropriate image is used to draw the button in response to system requests and in direct response to user interaction with the mouse. The CFastBmpBtn sends notification messages to the parent window in the same manner as standard buttons.
(2) The CFastBmpBtn::SetButtonFlags function allows an application to set the CFastBmpBtn::RestoreFocus flag for a particular button. When this flag is set, the button will remember which window previously had focus whenever it receives focus due to a mouse click, and will restore the focus to the window when the mouse button is released.
CFastBmpBtn is used to implement the graphical buttons in user interfaces for conferencing system 100. An example is the handset and control panel buttons in the conferencing application 502.
Data Conferencing Application
Data conferencing application 504 implements the data conferencing user interface. Data conferencing application is implemented as a Microsoft® Windows™ 3.1 application. The data conferencing application uses a "shared notebook" metaphor. The shared notebook lets the user copy a file from the computer into the notebook and review it with a remote user during a call. When the user is sharing the notebook (this time is called a "meeting"), the users see the same information on their computers, users can review it together, and make notes directly into the notebook. A copy of the original file is placed in the notebook, so the original remains unchanged. The notes users make during the meeting are saved with the copy in a meeting file. The shared notebook looks like a notebook or stack of paper. Conference participants have access to the same pages. Either participant can create a new page and fill it with information or make notes on an existing page. A preferred embodiment of a data conferencing application is d described i n U.S. patent application Ser. No. 08/137,319 (filed Oct. 14, 1993) and in U.S. patent application Ser. No. 08/170,146 (filed Dec. 20, 1993).
Conference Manager
Referring again to FIG. 5, audio/video conferencing application 502 supports audio and video conferencing between remote locations, while data conferencing application 504 supports the sharing of data (e.g., documents) between the remote locations. In general, conferencing system 100 is capable of simultaneously supporting multiple applications that support different types of conferencing services (e.g., audio/video conferencing, data sharing, and background file transfer).
When a single telephone line is used as the transport medium, the conference applications may need to share that line. Conference manager 544 (also known as CMDLL) coordinates connection and data channel activities for the conference applications. It provides capabilities to centralize and coordinate dial, hang-up, data channel management activities, and application launching. It enables conference applications to establish and tear down connections. It also provides applications access to already established connections. A connection is established through the conference manager 544 instead of calling the communication software (i.e., comm manager 518 via comm API 510) directly. Data channels are also obtained through the conference manager 544.
Conference manager 544 and conferencing API 506 provide the following advantages to conferencing system 100:
It is application aware (i.e., if application A on conferencing system X attempts to establish a data channel with application A on conferencing system Y, the conference manager 544 will automatically launch application A on system Y if application A is not already running).
It simplifies the establishment of a full duplex channel by providing a single simplified call to establish such a channel.
It allows applications that would normally use a single dedicated connection to share a connection.
It provides an efficient mechanism to inform applications about events such as "connection established" and "connection torn down."
It adds a layer of control for channel management (e.g., when an application with open channels terminates, the open channels are guaranteed to become closed).
The main purpose of conference manager 544 is to provide a set of services that allows several conference applications to share a common connection. The model is that once a connection is established by some application, any conference application can latch on to the connection and establish a full-duplex communication channel with its peer application running on the remote machine. The full duplex channel is implemented as a channel pair, or in TII terms, one outgoing and one incoming channel.
The conference manager services are used in conferencing system 100 to coordinate connection and data channel activities for the audio/video and data conferencing applications. The conference manager software sits between the applications (clients) and the communication software. A connection is established through the conference manager 544 instead of calling the communication software directly. Data channels are also obtained through the conference manager 544. Conference manager 544 also implements an application registry which gives it some intelligence as to which conference applications are running.
This approach has several advantages:
(1) Conference manager 544 is application aware. This means that if application A on computer X attempts to establish a data channel with application A on computer Y, conference manager 544 will automatically launch A on system Y if it is not already running.
(2) It simplifies the establishment of a full duplex channel. It provides a single simplified call to establish such a channel.
(3) It allows applications that would normally use a single dedicated connection to share a connection.
(4) It provides a nice mechanism to inform applications about events such as "connection established" and "connection torn down."
(5) It adds a layer of control for channel management. For example, when an application with open channels terminates, its channels are guaranteed to get closed.
Conference Manager Overview
Conference manager 544 consists of several modules. The most important ones are as follows:

__________________________________________________________________________cmcall.cContains the icmMakeCall and icmHangupCall procedures that arecalled fromthe CMIF library.cmchan.cContains the implementation of channel related procedures specifiedin theconference manager API. These are cmGetChannelPair,cmCloseChannelPair,and cmGetChannelPairInfo.cmclntfy.cContains dialog box procedures for the dialogs that are displayed onincomingcalls. There is one dialog for the case when the caller ID matches arecord inthe address book, and one dialog for the case when match is notfound.cmdll.cContains the LibMain and WEP procedures. Also contains variousinitialization procedures, including Lib.sub.-- InitializeCf whichloads VCI.DLL andmakes the VCI call CF.sub.-- Init to initialize the comm subsystemandLib.sub.-- TerminateCf which calls CF.sub.-- UnInit and unloadsVCI.DLL. This modulealso contains code for registering and unregistering with theaddress bookservices provided by ABDLL.DLL.cmmisc.cContains miscellaneous supporting functions used throughout theothermodules in CMDLL.cmnotify.cContains callbacks required by VCI. The callback Lib.sub.-- CfCallCallBackhandles the CFM.sub.-- * messages such as CFM.sub.-- CALL.sub.--NTFY andCFM.sub.-- CALL.sub.-- HANGUP. The callback Lib.sub.-- CfChanCallBack handles VCICFM.sub.-- CHAN.sub.-- * channel messages such as CFM.sub.--CHAN.sub.-- ACCEPT.sub.-- NTFY andCFM.sub.-- CHAN.sub.-- REJECT.sub.-- NTFY.cmreg.cContains the implementation of the conference manager API functionscmRegister, cmUnregister, and cmInstall.__________________________________________________________________________

Implementation Details
This section describes the implementation details of key areas of CMDLL 544.
Conference Application Installation
In order to make CMDLL 544 aware of conference applications, conference applications are preferably installed. Installed applications are listed in the Conference Apps! section in the PSVIDEO.INI file. Applications are typically installed directly by an installation program. It is also possible for an application to install itself by calling cmInstall (if, for example, the PSVIDEO.INI file has been corrupted subsequent to installation).
Conference Application Registration
Before a conference application makes use of CMDLL services, it is loaded and registered with the DLL. An application registers with CMDLL 544 by calling cmRegister. This function is in the module cmreg.c. CMDLL 544 keeps track of registered applications, and for each registered application, a CONFAPP structure is filled in. CMDLL 544 has a dynamically allocated array of CONFAPP structures. This array is built based on the applications that are installed (i.e., specified in the Conference Apps! sections in the PSVIDEO.INI file). If an application attempts to register without being installed cmRegister will fail.
After an application has registered with CMDLL 544, subsequent calls by said application do not require the application ID to be specified. CMDLL 544 keys off of the application's task handle and is able to map a task to an application ID Registered applications are notified through a callback about certain events such as connection establishment and connection tear-down.
VCI Call Handler Callback
CMDLL 544 is responsible for handling VCI 506 call callback messages. Most messages are generated by the comm subsystem as a result of calls to VCI 506. All calls in VCI 506 are asynchronous, hence the messages in this callback. This callback, Lib -- CfCallCallBack, is located in the module cmnotify.c and a pointer to the function is provided to VCI.DLL in the call CF -- Init. The Lib -- CfCallCallBack callback is defined as follows:
BOOL CALLBACK Lib -- CfCallCallBack(IN UINT uiMsg, IN WPARAM wParam, IN LPARAM lParam)
Channel Pair Establishment
CMDLL 544 provides a high-level service, cmGetChannelPair, that enables conference applications to establish easily a channel pair, i.e., one channel for inbound data and one channel for outbound data. The cmGetChannelPair uses VCI services (which in turn use TII services). Applications may establish up to 32 channel pairs. Each channel pair has an associated usage ID which is defined by the application. In this way, when an application establishes a channel pair, its peer (or peers in a multipoint environment) will know the purpose of channel pair. For example, one channel pair could be established for file transfer and another channel pair for control data. Appropriate channel handlers (e.g., TII/VCI) can thus be specified for different channels.
As noted earlier, CMDLL 544 keeps track of each application with an array of CONFAPP structures. A CONFAPP structure contains an array of structures of the type CHANNELPAIR, which is defined as follows:

______________________________________typedef struct tagCHANNELPAIRHCHAN hChanIn; // input (receive) channelHCHAN hChanOut; // output (send) channelWORD wState; // channel pair state (CPS.sub.-- *)CMCHAN.sub.-- INFO cmChanInfo; // channel info struct} CHANNELPAIR;______________________________________

In turn, each channel pair contains a CMCHAN -- INFO structure, which is defined as follows:

__________________________________________________________________________typedef struct tagCMCHAN.sub.-- INFOHCHAN hChanIn; // input (read) channelHCHAN hChanOut; // output (send) channelCHAN.sub.-- INFO chanInfo; // channel informationDWORD dwTransId; // transaction idBYTE byUsageId; // usage idBOOL bOpener; // TRUE if initiator of cmGetChannelPair, else FALSE} CMCHAN.sub.-- INFO;__________________________________________________________________________

This structure, in turn, contains the CHAN -- INFO structure defined by TII 510. When a channel pair is established, certain information is transferred between the peer CMDLLs. This information is transferred In the CHAN -- INFO structure. Successful channel pair establishment happens as follows. First a connection is established. The application on Site A calls cmGetChannelPair. CMDLL then handles all the VCI details of establishing outbound and inbound data channels using the CF -- OpenChannel and CF -- AcceptChannel VCI calls. Once the two channels have been established at the CMDLL level, CMDLL calls the applications back with the channel handles.
Once the application receives the channel handles through the CMN -- CHANNELPAIR message, the application registers a channel handler using the VCI call CF -- RegisterChanHandler. The cmGetChannelPair procedure fills in the Id field of the CHAN -- INFO structure and then calls CF -- OpenChannel. The rest of the processing for setting up the channel pairs takes place in the channel manager callback Lib -- CfChanCallBack. The Id field is important in that it identifies:
Which application is establishing a channel pair (which is important for CMDLL on the remote site so that it knows which application to notify).
The usage id for the channel pair (which is important for the remote application so that it knows what to do with the channel pair).
Whether the channel that is being opened is inbound or outbound.
Critical Sections
One of the key elements of CMDLL 544 is that it notifies conference applications of several events that take place in the comm subsystem, e.g., an incoming call. CMDLL 544 is also responsible for calling the comm subsystem in response to user-initiated events, e.g., hang up a call. CMDLL 544 is also responsible for starting and stopping playback of audio feedback through the PWave interface. CMDLL 544 may be interrupted by the comm subsystem while it is in the process of handling events initiated by the user. For example, while CMDLL 544 is in icmHangupCall processing, it may be interrupted by a CFM -- REJECT -- NTFY notification message from the comm subsystem. The critical section code prevents re-entrancy problems. It prevents the application from having to deal with call rejection messages when in fact it is already in the process of hanging up. Three global variables are declared in cmmisc.c for the purpose of critical sections:
UINT G -- nProgressCriticalSection=0;
UINT G -- nHangupCriticalSection=0;
UINT G -- nRejectCriticalSection=0;
These variables are manipulated and examined in cmcall.c in icmHangupCall and in the handling of the CFM -- REJECT -- NTFY and CFM -- PROGRESS -- NTFY messages in the VCI call callback routine Lib -- CfCallCallBack in cmnotify.c.
Call Notiification and Caller ID
CMDLL 544 is responsible for listening for incoming calls, notifying the user of incoming calls, and for rejecting or accepting the incoming call as specified by the user. On incoming calls, VCI 506 calls the Lib -- CfCallCallBack with the CFM -- CALL -- NTFY message. As outlined in the VCI Call Handler Callback section, if caller ID is available (through the IParam of the callback message), then the callback function performs a series of address book queries to determine the name and additional information of the caller.
Once the series of address book queries have been completed, the procedure CallNotifyDlg -- Do is called. It is responsible for calling one of two dialog box procedures: one if caller ID is unavailable from the comm subsystem or if the address book query failed, and a different dialog box if the address book query produced a match.
This procedure is also responsible for disabling all top-level windows (handset, remote, local, and snapshot). This is done to prevent the user from accessing other features when an incoming call is pending. Accessing other features when a call is pending causes re-entrancy problems.

______________________________________BOOL CallNotifyDlg.sub.-- Do(IN HWND hWndOwner, // owner of dialog windows (handset window)IN HAB.sub.-- REC hAbRec, // address book record, possibly NULLIN LPSTR lpszCallerID // caller ID string from comm subsystem______________________________________

Audible Call Progress
CMDLL 544 is responsible for providing audible call progress. CMDLL 544 uses the PWave services for starting and stopping playback of wave files. PWave can play a sound both synchronously and asynchronously. In the synchronous case, the number of times to play a file is specified, and the StartWavePlayback procedure does not return until it is finished playing the file the specified number of times. In the asynchronous case, the StartWavePlayback procedure returns immediately. In this case, PWave allows CMDLL to stop wave file playback at any time. Audible call progress is provided in the following situations:
Incoming call:
The RINGIN.WAV wave file starts playing asynchronously in the WM -- INITDIALOG case in the call notification dialog boxes. Playback ends when the user accepts the call, rejects the call. or the caller hangs up.
Incoming call in auto answer mode:
The AABEEP.WAV file is played once when a call comes in and the application is in auto-answer mode.
Outgoing call:
The RINGOUT.WAV file is played on the callers machine. This wave file is played to the caller to inform that the callee's machine is ringing. The RINGOUT.WAV file starts playing asynchronously when the comm subsystem calls the callback Lib -- CfCallCallBack with the CFM -- PROGRESS -- NTFY message (with LOWORD(lParam) equal to CF -- CALL -- RINGING). Playback stops if the caller hangs up, or the callee accepts or rejects the call.
Busy signal:
The BUSY.WAV file is played once on the caller's machine if the callee is already in a video conference.
Error signal:
The PROBLEM.WAV file is played once in response to the CFM -- REJECT -- NTFY message in the callback Lib -- CfCallCallBack.
The RINGIN.WAV file is the default wave file for incoming call notification. The user may optionally select a different wave file. This is done with a video preferences dialog box. If a different wave file has been selected, the PSVIDEO.INI file will contain the following entries:

______________________________________ Preferences!______________________________________AudioPreference=1 ; 0 means use defaultWavePath=c:\myring.wav ; if AudioPreference is 1, use this wave file for incoming calls______________________________________

The selected wave file meets the following criteria: sampled at 8 kHz, 8 bits, mono, and it is no larger than 64K.
On Demand Application Invocation
Conventional electronic conferencing applications require that both sites be running versions of the conferencing application prior to initiating the sharing of information (i.e., conferencing). As a result, users must confirm (e.g., via an independent telephone call) that the appropriate applications are running before sharing information.
In a conferencing network comprising preferred embodiments of conferencing system 100, only one site need be running a conferencing application before information sharing can be initiated. Moreover, if possible, the same application on the remote site is launched to complete the sharing. Conference manager 544 of FIG. 5 provides these capabilities. Conference manager 544 allows an application to install, register/unregister, make/hang-up calls, and establish/destroy communication channels. After successfully placing a call to a remote site, a conferencing application may try to establish a communication channel. In the process of establishing communication channels, the application is capable of being launched remotely if it is necessary. To accomplish this, all conferencing applications are assigned a unique application ID (i.e., APPID).
When an attempt to establish a communication channel is made, the application ID is used to identify the application for that channel. The conference manager 544 uses the APPID to determine (a) if the application is installed and (b) if the application is currently running. If the answer is yes to both of these questions, then the communication channel can be established immediately. If the answer is yes to (a) and no to (b), then the conference manager 544 is able to launch the desired application (via the Microsoft® WinExec function) and poll for registration. If the answer is no to both (a) and (b), then the communication channel will fail to be created.
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, according to a preferred embodiment of the present invention. On-demand application invocation applies when a conferencing application (App #1) running in one conferencing system (Site A) attempts to establish a conference with another conferencing system (Site B), where App #1 is installed but not currently running in Site B. App #1 in Site A starts the process by causing a request for a comm channel to be sent to Site B to establish communication between Site A and Site B. The comm channel request identifies the application running in Site A (e.g., APPID for App #1).
As shown in FIG. 33, the conference manager 544 of Site B receives the comm channel request from Site A (step 3302). The conference manager 544 of Site B retrieves the application ID for App #1 from the comm channel request and determines whether App #1 is installed in Site B (step 3304). If App #1 is not installed in Site B (step 3306), then the requested conference cannot proceed and the conference manager 544 of Site B causes the comm channel request of Site A to be rejected (step 3308).
Otherwise, if App #1 is installed in Site B (step 3306), then the conference manager 544 of Site B determines whether App #1 is registered, indicating that App #1 is already running in Site B. If App #1 is registered (step 3310), then processing continues to step 3318 as described below. Otherwise, if App #1 is not registered (step 3310), then the conference manager 544 of Site B attempts to synchronously launch App #1 (by calling the WinExec function of the Microsoft® Windows™ operating system) and thereby inform App #1 of Site B that a call is in progress (step 3312).
After attempting to launch App #1, the conference manager 544 of Site B checks to see whether App #1 was successfully launched by determining whether App #1 is now registered. If App #1 is still not registered (step 3314), then something went wrong in launching App #1 in Site B and again the conference manager 544 of Site B causes the comm channel request of Site A to be rejected (step 3316). Otherwise, if App #1 is now registered (step 3314), then the conference manager 544 of Site B accepts the comm channel request from Site A (step 3318) and notifies App #1 of Site B that the comm channel is open (step 3320) allowing conferencing to proceed.
The pseudocode for the local site communication channel establishment (Site A) is as follows:
request a communication channel
get notified when it has been established (or failed)
The pseudocode for the remote site communication channel establishment (Site B) is as follows:
get a communication channel request
get the appid from the request
if the application is not installed then
reject the communication channel request
elseif the application is not registered then
WinExec the application
inform the application there is a call in progress
if the application is still not registered then
reject the communication channel request
accept the communication channel request
notify application that comm channel is open
Through this on-demand invocation of applications, one conferencing system running a conferencing application can cause a remote conferencing system to invoke a corresponding application at the remote site. Those skilled in the art will understand that this capability alleviates the requirement for arranging for a conference by external means (e.g., via telephone) to coordinate the parallel independent launching of the corresponding conferencing applications in the remote sites.
Managing Multiple Applications
Comm API (i.e., transport independent interface (TII)) 510 of FIG. 5 establishes connections with remote conferencing systems for conferencing sessions. TII 510 also establishes one or more channels within each connection for use by conferencing applications (such as 502 and 504). These channels are used by the applications for transmitting or receiving different types of information with the remote conferencing systems. For example, audio/video conferencing application 502 uses four channels to transmit and receive audio and video signals to and from a remote conferencing system. Similarly, data conferencing application 504 uses two channels to transmit and receive data signals to and from a remote conferencing system.
The conference manager 544 of FIG. 5 provides the capability for two or more conferencing applications to share a single connection in a single conferencing session with a remote conferencing system. This capability allows two or more conferencing applications to participate in the same conferencing session simultaneously using a single connection between the local and remote conferencing systems.
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, according to a preferred embodiment of the present invention. The processing of FIG. 34 begins with the audio/video conferencing application 502 by asking the conference manager 544 to establish a connection for conferencing with a remote conferencing system (step 3402). Application 502 makes this request by calling the cmMakeCall function of the conference manager 544.
The conference manager 544 passes the connection request to the conferencing API (VCI) 506 by calling the CF -- MakeCall() function (step 3404). VCI 506 in turn passes the connection request to TII 510 by calling the MakeConnection function (step 3406). TII 510 causes the connection with the remote conferencing system to be established and also establishes four channels (i.e., transmit/receive audio/video) within that connection for the audio/video conferencing application 502 to use (step 3408). As part of this step, VCI 506 causes handles for the four channels to be passed back to application 502. TII 510 causes the connection and channels to be established by communicating with the peer TII 510 of the remote conferencing system.
Data conferencing application 504 then asks the conference manager 544 to establish channels within the established connection for transmitting and receiving data signals with the remote conferencing system (step 3410). Data conferencing application 504 knows that the connection has been established, because application 504 has already registered with the conference manager 544 and the conference manager 544 informs all registered applications of connections by sending the CMN -- CALL message. Since data conferencing application 504 already knows that the connection has been established, application 504 makes the channel request by calling the cmGetChannelPair function of the conference manager 544.
The conference manager 544 then passes the channel request to the VCI 506 (by calling CF -- OpenChannel) (step 3412). VCI 506 in turn passes the channel request to TII 510 (by calling OpenChannel) (step 3414). Conference manager 544 establishes the two requested channels for data conferencing application 504 within the already established connection with the remote conferencing system (step 3416). As part of this step. conference manager 544 causes handles for the two channels to be passed back to application 504.
The conferencing session is then able to proceed with both applications 502 and 504 using a single connection with the remote conferencing system for its different channels (step 3418).
Conferencing API
Conferencing API 506 of FIG. 5 (also known as video conferencing interface (VCI)) facilitates the easy implementation of conferencing applications 502 and 504. Conferencing API 506 of FIG. 5 provides a generic conferencing interface between conferencing applications 502 and 504 and the video, comm, and audio subsystems. Conferencing API 506 provides a high-level abstraction of the services that individual subsystems (i.e., video audio, and comm) support. The major services include:
Making, accepting, and hanging-up calls.
Mediating conference requirements between peers.
Establishing and terminating multiple communication channels for individual subsystems.
Instantiating and controlling local video and audio.
Sending video and audio to a remote site through the network.
Receiving, displaying, and controlling the remote video and audio streams.
Conferencing applications 502 and 504 can access these services through the high-level conferencing API 506 without worrying about the complexities of low-level interfaces supported in the individual subsystems.
In addition, conferencing API 506 facilitates the integration of individual software components. It minimizes the interactions between conferencing applications 502 and 504 and the video, audio, and comm subsystems. This allows the individual software components to be developed and tested independent of each other. Conferencing API 506 serves as an integration point that glues different software components together. Conferencing API 506 facilitates the portability of audio/video conferencing application 502.
Conferencing API 506 is implemented as a Microsoft® Windows™ Dynamic Link Library (DLL). Conferencing API 506 translates the function calls from conferencing application 502 to the more complicated calls to the individual subsystems (i.e., video, audio, and comm). The subsystem call layers (i.e., video API 508, comm API 510, and audio API 512) are also implemented in DLLs. As a result, the programming of conferencing API 506 is simplified in that conferencing API 506 does not need to implement more complicated schemes, such as dynamic data exchange (DDE), to interface with other application threads that implement the services for individual subsystems. For example, the video subsystem will use window threads to transmit/receive streams of video to/from the network.
Conferencing API 506 is the central control point for supporting communication channel management (i.e., establishing, terminating channels) for video and audio subsystems. Audio/video conferencing application 502 is responsible for supporting communication channel management for the data conferencing streams.
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, video API 510, and audio API 512, on the other side, according to a preferred embodiment of the present invention. Conferencing API 506 comprises conferencing finite state machine (FSM) 702, conferencing primitive validator 704, conferencing primitive dispatcher 708, conferencing callback 706, comm primitive 712, comm callback 710, video primitive 716, and audio primitive 720 of FIG. 7.
Conferencing primitive validator 704 validates the syntax (e.g., checks the conferencing call state, channel state. and the stream state with the conferencing finite state machine (FSM) 702 table and verifies the correctness of individual parameters) of each API call. If an error is detected, primitive validator 704 terminates the call and returns the error to the application immediately. Otherwise, primitive validator 704 calls conferencing primitive dispatcher 708, which determines which subsystem primitives to invoke next.
Conferencing primitive dispatcher 708 dispatches and executes the next conferencing API primitive to start or continue to carry out the service requested by the application. Primitive dispatcher 708 may be invoked either directly from primitive validator 704 (i.e., to start the first of a set of conferencing API primitives) or from conferencing callback 706 to continue the unfinished processing (for asynchronous API calls).
After collecting and analyzing the completion status from each subsystem, primitive dispatcher 708 either (1) returns the concluded message back to the conferencing application by returning a message or invoking the application-provided callback routine or (2) continues to invoke another primitive to continue the unfinished processing.
There are a set of primitives (i.e., comm primitives 712, video primitives 716, and audio primitives 720) implemented for each API call. Some primitives are designed to be invoked from a callback routine to carry out the asynchronous services.
The subsystem callback routine (i.e., comm callback 710) returns the completion status of an asynchronous call to the comm subsystem to conferencing callback 706, which will conduct analysis to determine the proper action to take next. The comm callback 710 is implemented as a separate thread of execution (vthread.exe) that receives the callback Microsoft® Windows™ messages from the comm manager and then calls VCI DLL to handle these messages.
Conferencing callback 706 returns the completion status of an asynchronous call to the application. Conferencing callback 706 checks the current message/event type, analyzes the type against the current conferencing API state and the next primitive being scheduled to determine the actions to take (e.g., invoke another primitive or return the message to the application). If the processing is not complete yet, conferencing callback 706 selects another primitive to continue the rest of the processing. Otherwise, conferencing callback 706 returns the completion status to the application. The conferencing callback 706 is used only for comm related conferencing API functions; all other conferencing API functions are synchronous.
The major services supported by conferencing API 506 are categorized as follows:
Initialization and Call Services (establish/terminate a conference call).
Stream Services (capture, play, record, link, control the multimedia audio and video streams, and access and manipulate data from the streams).
Channel Services (establish/terminate channels on the call, and send/receive data on the channels).
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 BeginSession to initialize their comm subsystems.
Conferencing APIs A and B enter a PeekMessage loop waiting for a SESS -- BEGIN callback from the comm subsystem.
Uninitialization of a session in the comm subsystem is implemented as follows:
Conferencing APIs A and B call EndSession to uninitialize their comm subsystems.
Conferencing APIs A and B receive a SESS -- CLOSED callback from the comm subsystem.
Conferencing APIs A and B then notify the conferencing applications with a CFM -- UNINIT -- 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 MakeConnection to dial conferencing API B's number.
Conferencing API B receives a CONN -- REQUESTED callback from the comm subsystem.
Conferencing API B sends the call notification to the graphic user interface (GUI) with a CFM -- CALL -- NTFY callback; and if user B accepts the call via the GUI, conferencing API B proceeds with the following steps.
Conferencing API B calls AcceptConnection to accept the incoming call from conferencing API A.
Conferencing APIs A and B receive CONN -- ACCEPTED callback from the comm subsystem.
Conferencing API A calls OpenChannel to open its outgoing conferencing control channel.
Conferencing API B receives the CHAN -- 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 -- 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 -- REQUESTED callback for the incoming control channel and accepts it via AcceptChannel.
Conferencing API B receives the CHAN -- 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 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.
Conferencing API A calls OpenChannel to open its outgoing audio channel.
Conferencing API B receives the CHAN -- REQUESTED callback for the incoming audio channel and accepts it via AcceptChannel.
Conferencing API A receives the CHAN -- 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 then notify the conferencing applications with a CFM -- ACCEPT -- 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 then calls the comm manager's CloseConnection.
The comm manager implicitly closes all channels, and sends Chan -- Closed callbacks to conferencing API A.
Conferencing API A closes its remote audio/video streams on receipt of the Chan -- Closed callback for its inbound audio/video channels, respectively.
Conferencing API A then receives the CONN -- CLOSE -- RESP from the comm manager after the call is cleaned up completely. Conferencing API A notifies its application via a CFM -- HANGUP -- NTFY.
In the meantime, the comm manager on B would have received the hangup notification, and would have closed its end of all the channels, and notified conferencing API B via Chan -- Closed.
Conferencing API B closes its remote audio/video streams on receipt of the Chan -- 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 -- Closed callback for its outbound audio/video channels, respectively.
Conferencing API B then receives a CONN -- CLOSED notification from its comm manager. Conferencing API B notifies its application via CFM -- HANGUP -- NTFY.
Interfacing with the Audio and Video Subsystems
Conferencing API 506 supports the following services with the audio and video subsystems:
Capture/monitor/transmit local video streams.
Capture/transmit local audio streams.
Receive/play remote streams.
Control local/remote streams.
Snap an image from local video stream.
Since the video and audio streams are closely synchronized, the audio and video subsystem services are described together.
Capture/Monitor/Transmit Local Streams
The local video and audio streams are captured and monitored as follows:
Call AOpen to open the local audio stream.
Call VOpen to open the local video stream.
Call ACapture to capture the local audio stream from the local hardware.
Call VCapture to capture the local video stream from the local hardware.
Call VMonitor to monitor the local video stream.
The local video and audio streams are begun to be sent out to the remote site as follows:
Call ALinkOut to connect the local audio stream to an output network channel.
Call VLinkOut to connect the local video stream to an output network channel. The monitoring of the local video stream locally is stopped as follows:
Call VMonitor(off) to stop monitoring the local video stream.
Receive/Play Remote Streams
Remote streams are received from the network and played as follows:
Call AOpen to open the local audio stream.
Call VOpen to open the local video stream.
Call ALinkIn to connect the local audio stream to an input network channel.
Call VLinkIn to connect the local video stream to an input network channel.
Call APlay to play the received remote audio stream.
Call VPlay to play the received remote video stream.
Control Local/Remote Streams
The local video and audio streams are paused as follows:
Call VLinkout(off) to stop sending local video on the network.
Call AMute to stop sending local audio on the network.
The remote video and audio streams are paused as follows:
If CF -- PlayStream(off) is called, conferencing API calls APlay(off) and VPlay(off).
The local/remote video/audio streams are controlled as follows:
Call ACntl to control the gains of a local audio stream or the volume of the remote audio stream.
Call VCntl to control such parameters as the brightness, tint, contrast, color of a local or remote video stream.
Snap an Image from Local Video Streams
A snapshot of the local video stream is taken and returned as an image to the application as follows:
Call VGrabframe to grab the most current image from the local video stream.
Conferencing API 506 supports the following function calls by conferencing applications 502 and 504 to the video comm, and audio subsystems:

__________________________________________________________________________CF.sub.-- Init Reads in the conferencing configuration parameters from an initialization file; loads and initializes the software of the comm, video, and audio subsystems by allocating and building internal data structures; allows the application to choose between the message and the callback routines to return the event notifications from the remote site.CF.sub.-- MakeCall Makes a call to the remote site to establish a connection for conferencing. The call is performed asynchronously.CF.sub.-- AcceptCall Accepts call initiated from the remote site based on information received in the CFM.sub.-- CALL.sub.-- NTFY message as delivered to the graphical user interface.CF.sub.-- RejectCall Rejects incoming call, if appropriate, upon receiving a CFM.sub.-- CALL.sub.-- NTFY message as delivered to the GUI.CF.sub.-- HangupCall Hangs up a call that was previously established; releases all resources, including all types of streams and data structures, allocated during the call.CF.sub.-- GetCallInfo Returns the information about the specified call, including its current state.CF.sub.-- CapMon Starts the capture of analog video signals from the local camera and displays the video in the local video window which is pre-opened by the application. This function allows the user to preview his/her appearance before sending the signals out to the remote site.CF.sub.-- PlayRcvd Starts the reception and display of remote video signals in the remote video window, which is pre-opened by the application; starts the reception and play of remote audio signals through the local speaker.CF.sub.-- DestroyStream Destroys the specified stream group that was created by CF.sub.-- CapMon or CF.sub.-- PlayRcvd. As part of the destroy process, all operations (e.g., sending/playing) being performed on the stream group will be stopped and all allocated system resources will be freed.CF.sub.-- Mute Uses AMute to turn on/off the mute function being performed on the audio stream of a specified stream group. This function will temporarily stop or restart the related operations, including playing and sending, being performed on this stream group. This function may be used to hold temporarily one audio stream and provide more bandwidth for other streams to use.CF.sub.-- SnapStream Takes a snapshot of the video stream of the specified stream group and returns a still image (reference) frame to the application buffers indicated by the hBuffer handle.CF.sub.-- ControlStream Controls the capture or playback functions of the local or remote video and audio stream groups.CF.sub.-- SendStream Uses ALinkOut to pause/unpause audio.CF.sub.-- GetStreamInfo Returns the current state and the audio video control block (AVCB) data structure, preallocated by the application, of the specified stream groups.CF.sub.-- PlayStream Stops/starts the playback of the remote audio/video streams by calling APlay/VPlay.CF.sub.-- GetAudVidStream Returns the audio and video stream handles for the specified stream group.CF.sub.-- RegisterChanMgr Registers a callback or an application window whose message processing function will handle notifications generated by network channel initialization operations. This function is invoked before any CF.sub.-- OpenChannel calls are made.CF.sub.-- OpenChannel Requests to open a network channel with the peer application. The result of the action is given to the application by invoking the callback routine specified by the call to CF.sub.-- RegisterChanMgr. The application specifies an ID for this transaction. This ID is passed to the callback routine or posted in a message.CF.sub.-- AcceptChannel A peer application can issue CF.sub.-- AcceptChannel in response to a CFM.sub.-- CHAN.sub.-- OPEN.sub.-- NTFY message that has been received. The result of the CF.sub.-- AcceptChannel call is a one-way network channel for receiving data.CF.sub.-- RejectChannel This routine rejects a CFM.sub.-- CHAN.sub.-- OPEN.sub.-- NTFY from the peer.CF.sub.-- RegisterChanHandler This function registers a callback or an application window whose message processing function will handle notifications generated by TII network channel IO activities. The channels that are opened will receive TII CHAN.sub.-- DATA.sub.-- SENT notifications, and the channels that are accepted will receive TII CHAN.sub.-- RCV.sub.-- COMPLETE notifications.CF.sub.-- CloseChannel This routine will close a network channel that was opened by CF.sub.-- AcceptChannel or CF.sub.-- OpenChannel. The handler for this channel is automatically de-registered.CF.sub.-- SendData Send data to peer. If the channel is not reliable and there are no receive buffers posted on the peer machine, the data will be lost.CF.sub.-- RecvData Data is received through this mechanism. Normally this call is issued in order to post receive buffers to the system. When the system has received data in the given buffers, the Channel Handler will receive the TII CHAN.sub.-- RCV.sub.-- COMPLETE notification.CF.sub.-- GetChanInfo This function will return various statistical information about a channel. For example: bandwidth information, number of sends/second, number of receives/second, etc.__________________________________________________________________________

These functions are defined in further detail later in APPENDIX A entitled "Conference Manager API."
In addition, conferencing API 506 supports the following 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.-- CALL.sub.-- NTFY Indicates that a call request initiated from the remote site has been received.CFM.sub.-- PROGRESS.sub.-- NTFY Indicates that a call state/progress notification has been received from the local phone system support.CFM.sub.-- ACCEPT.sub.-- NTFY Indicates that the remote site has accepted the call request issued locally. Also sent to the accepting application when CF.sub.-- AcceptCall completes.CFM.sub.-- REJECT.sub.-- NTFY Indicates that the remote site has rejected or the local site has failed to make the call.CFM.sub.-- HANGUP.sub.-- NTFY Indicates that the local or remote site has hung up the call.CFM.sub.-- UNINIT.sub.-- NTFY Indicates that uninitialization of comm subsystem has completed.CFM.sub.-- ERROR.sub.-- NTFY Indicates that a SESS.sub.-- ERROR was received from comm subsystem.__________________________________________________________________________

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), according to a preferred embodiment of the present invention. The possible conferencing call states are as follows:

__________________________________________________________________________CCST.sub.-- NULL Null State - state of uninitialized caller/callee.CCST.sub.-- IDLE Idle State - state of caller/callee ready to make/receive calls.CCST.sub.-- CALLING Calling state - state of caller trying to call callee.CCST.sub.-- CALLED Called state - state of callee being called by caller.CCST.sub.-- ACCEPTING Accepting state - state of accepting call from caller.CCST.sub.-- CONNECTED Call state - state of caller and callee during conferencing session.CCST.sub.-- CLOSING A hangup or call cleanup is in progress.__________________________________________________________________________

At the CCST -- CONNECTED state, the local application may begin capturing, monitoring, and/or sending the local audio/video signals to the remote application. At the same time, the local application may be receiving and playing the remote audio/video signals.
Referring now to FIG. 9, there is shown a representation of the conferencing stream FSM for each conferencing system participating in a conferencing session, according to a preferred embodiment of the present invention. The possible conferencing stream states are as follows:

__________________________________________________________________________CSST.sub.-- INIT Initialization state - state of local and remote streams after CCST.sub.-- CONNECTED state is first reached.CSST.sub.-- ACTIVE Capture state - state of local stream being captured. Receive state - state of remote stream being received.CSST.sub.-- FAILURE Fail state - state of local/remote stream after resource__________________________________________________________________________ failure.

Conferencing stream FSM represents the states of both the local and remote streams of each conferencing system. Note that the local stream for one conferencing system is the remote stream for the other conferencing system.
In a typical conferencing session between a caller and a callee, both the caller and callee begin in the CCST -- NULL call state of FIG. 8. The conferencing session is initiated by both the caller and callee calling the function CF -- Init to initialize their own conferencing systems. Initialization involves initializing internal data structures, initializing communication and configuration information, and verifying the local user's identity. The CF -- Init function takes both the caller and callee from the CCST -- NULL call state to the CCST -- IDLE call state. The CF -- Init function also places both the local and remote streams of both the caller and callee in the CSST -- INIT stream state of FIG. 9.
Both the caller and callee call the CF -- CapMon function to start capturing local video and audio signals and playing them locally, taking both the caller and callee local stream from the CSST -- INIT stream state to the CSST -- ACTIVE stream state. Both the caller and callee may then call the CF -- ControlStream function to control the local video and audio signals, leaving all states unchanged.
The caller then calls the CF -- MakeCall function to initiate a call to the callee, taking the caller from the CCST -- IDLE call state to the CCST -- CALLING call state. The callee receives and processes a CFM -- CALL -- NTFY message indicating that a call has been placed from the caller, taking the callee from the CCST -- IDLE call state to the CCST -- CALLED call state. The callee calls the CF -- AcceptCall function to accept the call from the caller, taking the callee from the CCST -- CALLED call state to the CCST -- ACCEPTING call state. The caller and callee receive and process a CFM -- ACCEPT -- NTFY message indicating that the callee accepted the call, taking the caller and callee from the CCST -- CALLING/CCST -- ACCEPTING call states to the CCST -- CONNECTED call state.
Both the caller and callee then call the CF -- PlayRcvd function to begin reception and play of the video and audio streams from the remote site, leaving all states unchanged. Both the caller and callee call the CF -- SendStream function to start sending the locally captured video and audio streams to the remote site, leaving all states unchanged. If necessary, both the caller and callee may then call the CF -- ControlStream function to control the remote video and audio streams, again leaving all states unchanged. The conferencing session then proceeds with no changes to the call and stream states. During the conferencing session, the application may call CF -- Mute, CF -- PlayStream, or CF -- SendStream. These affect the state of the streams in the audio/video managers, but not the state of the stream group.
When the conferencing session is to be terminated, the caller calls the CF -- HangupCall function to end the conferencing session, taking the caller from the CCST -- CONNECTED call state to the CCST -- IDLE call state. The callee receives and processes a CFM -- HANGUP -- NTFY message from the caller indicating that the caller has hung up, taking the callee from the CCST -- CONNECTED call state to the CCST -- IDLE call state.
Both the caller and callee call the CF -- DestroyStream function to stop playing the remote video and audio signals, taking both the caller and callee remote streams from the CSST -- ACTIVE stream state to the CSST -- INIT stream state. Both the caller and callee also call the CF -- DestroyStream function to stop capturing the local video and audio signals, taking both the caller and callee local streams from the CSST -- ACTIVE stream state to the CSST -- INIT stream state.
This described scenario is just one possible scenario. Those skilled in the art will understand that other scenarios may be constructed using the following additional functions and state transitions:
If the callee does not answer within a specified time period, the caller automatically calls the CF -- HangupCall function to hang up, taking the caller from the CCST -- CALLING call state to the CCST -- IDLE call state.
The callee calls the CF -- RejectCall function to reject a call from the caller, taking the callee from the CCST -- CALLED call state to the CCST -- IDLE call state. The caller then receives and processes a CFM -- REJECT -- NTFY message indicating that the callee has rejected the caller's call, taking the caller from the CCST -- CALLING call state to the CCST -- IDLE call state.
The callee (rather than the caller) calls the CF -- HangupCall function to hang up, taking the callee from the CCST -- CONNECTED call state to the CCST -- IDLE call state. The caller receives a CFM -- HANGUP -- NTFY message from the callee indicating that the callee has hung up, taking the caller from the CCST -- CONNECTED call state to the CCST -- IDLE call state.
The CF -- GetCallInfo function may be called by either the caller or the callee from any call state to determine the current call state without changing the call state.
During a conferencing session, an unrecoverable resource failure may occur in the local stream of either the caller or the callee causing the local stream to be lost, taking the local stream from the CSST -- ACTIVE stream state to the CSST -- FAILURE stream state. Similarly, an unrecoverable resource failure may occur in the remote stream of either the caller or the callee causing the remote stream to be lost, taking the remote stream from the CSST -- ACTIVE stream state to the CSST -- FAILURE stream state. In either case, the local site calls the CF -- DestroyStream function to recover from the failure, taking the failed stream from the CSST -- FAILURE stream state to the CSST -- INIT stream state.
The CF -- GetStreamInfo function may be called by the application from any stream state of either the local stream or the remote stream to determine information regarding the specified stream groups. The CF -- SnapStream function may be called by the application for the local stream in the CSST -- ACTIVE stream state. All of the functions described in this paragraph leave the stream state unchanged.
Video Subsystem
The video subsystem of conferencing system 100 of FIG. 5 comprises video API 508, video manager 516, video capture driver 522, video decode driver 548, video playback driver 550, and video/host interface 526 running on host processor 202 of FIG. 2 and video microcode 530 running on video board 204. The following sections describe each of these constituents of the video subsystem.
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, according to a preferred embodiment of the present invention. 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. One difference is the addition of the following function calls supported by video API 508:

______________________________________VReOpen Re-opens a video stream.VReportStats Grabs the statistics which are being updated periodically.______________________________________

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, according to a preferred embodiment of the present invention. 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.
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.
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.
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. 08/157,694. In addition, the time period δ 1 is a configuration-defined value with a preferred default setting of 400 milliseconds. Experiments indicate that, at 10-15 fps, if the audio and video streams drift by greater then 400 milliseconds, then a loss of lip synchronization is readily detected by the user.
The time period δ 2 may be defined based on the following criteria:
(1) It is made a function of δ 1 , so as to reduce the number of configuration parameters.
(2) It is large enough that the network transfer time of a restart key frame does not cause the restart sequence (key frame and follow-on delta frame) to be way-out-of-sync. Assuming IRV key frames of approx. 6 KB. and a network transfer time of 84 kbps, a key frame transfer takes approximately 600 msecs. Given the first criterion, δ 1 should be selected carefully. The follow-on delta frame is generated (and time-stamped for lip-sync) after a bit-rate throttle of the IRV capture driver to account for key frame size, and this throttle is concurrent with the key frame transmission.
(3) It is large enough that once audio and video have drifted to such an extent, a stream can not be effectively re-synchronized with the use of codec hurry-up logic; i.e., that it is more expedient to re-sync with a key frame (restart stream).
In a preferred embodiment, δ 2 =2*δ 1 .
In a preferred embodiment, the capture latency is not used and the playback latency default is 200 milliseconds. This value was experimentally determined. In fact, because this is a latency adjustment that corrects for the video capture interface to the host-to-network interface path, it may be a function of host performance, and as such should be derived as a function of the host calibration process.
Codec Hurry-up is also used when the playback display frequency is lowered. The video manager has the option, in order to decrease host utilization, of scaling the display rate. For example, only displaying every other, every third, etc. frame. If a display frequency less than one is specified, codec Hurry-up is used, as the stream does not require a color conversion for purposes of display.
Alternative Timestamp Driver
In a typical video conferencing scenario, the video frames generated at local conferencing system A will be time-stamped according to the capture clock at system A. The video frames will then be transmitted over a network to remote conferencing system B, which will maintain a playback clock. Although it is desirable to maintain high-resolution clocks (e.g., 32-bit time values) at both sites, the limited bandwidth of the communications network between the two sites may make transmission of entire clock values impractical. To satisfy both goals, remote system B preferably maintains its own version of the capture clock of local system A in a high-resolution counter in system B. This high-resolution counter is maintained by system B with information received from system A, where the information comprises a subset of the bits that represent each high-resolution clock value. This process is described in the context of the example depicted in FIG. 35.
At step 3502 of FIG. 35, the 12-bit (i.e., high-resolution) counters of systems A and B are synchronized. In step 3504, system A increments its 12-bit counter and sends the 4 low-order bits (e.g., as a time stamp for a video frame) to system B. In step 3506, system B adjusts its 12-bit counter based on the 4-bit value received from system A. At step 3506, systems A and B are again synchronized. In step 3508, system A again increments its 12-bit counter and sends the appropriate 4-bit time stamp to system B. In step 3510, when system B receives a 4-bit value of (0000), it recognizes that the counter has rolled over. In response, system B adds one to its counter causing it to carry a one into the high-order bits (i.e., the non-transmitted bits). Once again, system A and B are synchronized after step 3510.
Step 3512 demonstrates an error condition. System B is able to detect an error condition, because it knows what value to expect next. When it receives a value other than what it expects, system B recognizes that values have been dropped. In order to recover, system B updates its counter value using the received value, taking into account any required carry into the high-order bits.
Error conditions are a primary reason for transmitting more than one bit in the time stamp. Enough bits are preferably transmitted to ensure that at least one transmitted packet is received before a bit pattern is sent twice. If it can be guaranteed that the transmission is reliable, then only one bit is needed to keep remote versions of a counter of any resolution.
The process of FIG. 35 may be generalized for any two distinct memory areas that need to track a single incrementing or decrementing number across some limited bandwidth medium. Those skilled in the art will understand that the process may be generalized to clocks and time stamps of different sizes.
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. 08/157,694. In addition, a throttle rate is the percentage rate at which a bit stream data rate will be modified. For example, a throttle rate of 10 will decrease or increase a bit rate by 10%. A throttle down rate is a throttle rate for decreasing a bit stream data rate, and a throttle up rate is a throttle rate for increasing a bit stream data rate. A bit rate filter, for the purposes of this specification, is a function which takes as input a number of bit rate values (typically, the bit rates from some number of previous samples) and outputs a single bit rate. The specifics of the filter implementation are provided below.
The bit rate throttling algorithm is described as follows:
Immediately after posting N bits to the network for transmission:

______________________________________bits.sub.-- outstanding += N;if (bits.sub.-- outstanding > bit rate filter*throttle down latency)thencurrent.sub.-- bit.sub.-- rate = current.sub.-- bit.sub.-- rate*throttleup rate<program the capture driver with current.sub.-- bit.sub.-- rate>end.if______________________________________

Immediately after the completion of transmission of N bits on the network:

______________________________________bits.sub.-- outstanding = N;if (bits.sub.-- outstanding < bit rate filter*throttle up latency) thencurrent.sub.-- bit.sub.-- rate += current.sub.-- bit.sub.-- rate*throttleup rate<program the capture driver with current.sub.-- bit rate>end if______________________________________

The bit rate filter, for frame N, referred to as brN is defined as follows:

br.sub.N =(br.sub.N-1 +br.sub.N-2)/2,

where br 0 =br 1 =default bit rate.
Multiple Video Formats
Conferencing system 100 presents both a local monitor display and a remote playback display to the user. A digital video resolution of (160×120) is preferably used as capture resolution for ISDN-based video conferencing (i.e., the resolution of a coded compressed video stream to a remote site). (160×120) and (320×240) are preferably used as the local monitor display resolution. (320×240) resolution may also be used for high-resolution still images. Generating the local monitor display by decompressing and color converting the compressed video stream would be computationally expensive. The video capture driver 522 of FIG. 5 simultaneously generates both a compressed video stream and an uncompressed video stream. Video manager 516 makes use of the uncompressed video stream to generate the local monitor display. Video manager 516 may select the format of the uncompressed video stream to be either YUV-9 or 8-bits/pixel (bpp) RGB--Device Independent Bitmap (DIB) format. For a (160×120) local monitor, the uncompressed DIB video stream may be displayed directly. For a (320×240) monitor, a (160×120) YUV-9 format is used and the display driver "doubles" the image size to (320×240) as part of the color conversion process.
In the RGB and YUV-9 capture modes, RGB or YUV data are appended to capture driver IRV buffers, so that the capture application (VCapt EXE 1104) has access to both fully encoded IRV frames and either kGB or YUV data. Conferencing system 100 has custom capture driver interfaces to select either RGB capture mode, YUV capture mode, or neither.
Consider the overhead of the playback subsystem, assuming all video frames are displayed (full frame rate).
(1) A frame is decompressed using inter- and intra-frame decompression algorithms.
(2) The frame is color-converted from the color space of the decompressed stream to the color space of the display subsystem; e.g., from YUV to RGB.
(3) The frame is displayed by copying it to a display memory (e.g., via Microsoft® Windows™ GDI).
The first and second steps may be eliminated if RGB data is used as a video format. The first step may be eliminated if YUV data is used as a video format. These are important savings, in that step (2) may cost as much (or almost as much) as the first step and, depending on the graphics subsystem, the third step may be small compared to steps (1) and (2).
The technique of sending uncompressed video, in an RGB or YUV format, to the host for local monitoring, in order to eliminate steps (1) and/or (2), is referred to herein as Looking Glass. RGB or YUV data is used for the local monitor only, and is typically not appropriate for network transmission. To make for an efficient implementation, Looking Glass RGB or YUV frames are piggybacked behind the IRV stream, so that the number of frame complete callbacks is not increased at the host.
(160×120) resolution is used as the capture resolution for ISDN-based video conferencing; i.e., it is the resolution of a coded compressed video stream to a remote site. (160×120) is also used as a display resolution, and is referred to herein as the "normal" resolution. (320×240) resolution is used to provide a "quarter" screen (relative to 640×480 VGA) video display (and as a capture resolution for high-quality still images).
Normal Display Resolution
At 10 fps, a (160×120) 8-bits/pixel (bpp) capture resolution generates approximately 1.5 Mbits/sec (=192 KBytes/sec) data stream. For video conferencing, the (160×120) uncompressed (8 or 9 bpp) format is not excessive for ISA bus 208 of FIG. 2. Given the benefits of eliminating an IRV decode by host processor 202, as described above, it is advantageous to implement a local monitor using an uncompressed (160×120) video stream.
Because VRAM 304 of video board 204 of FIG. 3 contains the current decoded frame (used to produce the next encoded delta frame), uncompressed video frames are available to the host for local monitoring. In addition, the video board 204 has the capability to color convert (i.e., YUV to RGB) the decoded, uncompressed frame, as needed. For a (160×120) monitor, an RGB (palletized, 8 bpp) format is preferably used. Using RGB eliminates all decode and color conversion overhead at playback.
Consider a specific example. Assume the host processor 202 of FIG. 2 is an Intel® 486DX2™ 66 MHz processor. IRV decode ranges, based on the difficulty (amount of motion) of the video, from approximately 12-24 milliseconds. On the other hand, the transfer rate of ISA bus 208 is approximately 2 Mbytes/sec. A (160×120) RGB frame is (160×120)=19.2 KBytes, so its transfer takes approximately 10 msecs. Thus, using RGB data saves the host from 2-14 msecs/frame. At 10 fps, this is a 2-14% utilization improvement. Slower host processors show a more substantial improvement.
Quarter Display Resolution
At 10 fps, a quarter-display (320×240) 8-bits/pixel capture resolution generates four times the data, or approximately 6 Mbits/sec (=768 MBytes/sec), as the normal case. For video conferencing, the (320×240) un-encoded (8 or 9 bpp) format is considered excessive for ISA bus 208. Thus, uncompressed video frame data is preferably transferred using at most a (160×120) resolution.
One option for achieving quarter-display resolution is to use the Microsoft® GDI StretchBlt function. In many cases, however using the StretchBlt function, from a normal to quarter resolution, is expensive. As a result, using the GDI to achieve a (320×240) display at 10 fps is not generally feasible.
Another option is to perform special-purpose codec "doubling" from normal to quarter resolution. Since, in a preferred embodiment of the present invention, there is no RGB interface to the codec for the purposes of the special-purpose doubling feature, doubling is supported at the YUV interface only. An (160×120) color convert with "doubling" to (320×240) using an Intel® 486DX2 66 MHz processor takes approximately 8 msecs. A color convert with doubling plus quarter bit is generally much faster than a GDI normal to quarter stretch.
Thus, for a quarter screen (320×240) monitor, a (160×120) YUV-9 Looking Glass format is used, and the codec "doubles" the data. (160×120) YUV-9 also generates approximately 192 KBytes/sec (10 fps). Using YUV-9 data means that the IRV decode step of the playback can be eliminated.
Video Frame Format/Capture Interface
The format of the IRV/Looking Glass piggy-backed data is defined as follows:

______________________________________lpData Long pointer to the video frame data buffer.dwBufferLength Length of the data buffer pointed to by lpData, in bytes.dwBytesUsed Length of bytes used in the data buffer.dwTimeCaptured Time, in milliseconds, between the current frame and the beginning of the capture session.dwUser Reserved for application use.dwFlags Information about the data buffer.dwReserved 0! Reserved.dwLGBytesUsed Number of bytes used for Looking Glass data.dwReserved 2! Reserved.dwReserved 3! Reserved.bIRVData ! Compressed IRV data for video frame.bLGData ! Looking Glass data for video frame.______________________________________

The Microsoft® VideoHdr structure is augmented as follows:
(1) lpData references a combined IRV and Looking Glass (RGB or YUV) data array. If Looking Glass data is present, then it immediately follows (on the next byte boundary) the IRV data. Looking Glass data is either in an RGB DIB 8 bpp format, or YUV9 format. The YUV9 format can be used as input the IRV codec for purposes of color conversion (with optional doubling).
(2) If Looking Glass data is present, the dwReserved 1! is non-zero and equals the number of Looking Glass RGB or YUV bytes.
(3) If dwReserved 1! is non-zero, then VideoHdr->lpData+VideoHdr->dwBytesUsed references the first byte of the Looking Glass data.
Playback Implementation
Given that Looking Glass data may be present for monitor playback and IRV data is used for remote playback, the playback application, VPlay, must be aware of Looking Glass usage and take a different path of monitor and remote playback. VPlay executes the following cases for monitor playback (from within the ProcessPlayFrame proc):

______________________________________if capture format is IRV thenif Looking GIass is disabled thenDecompress Frameelse //Looking Glass is enabled.if display window size is quarter thenColorConvertFrame // YUV data.elseFastCopy // RGB data copied directly to outputdisplay buffer.else // YUV snapshot capture format.. . .______________________________________

Referring again to FIG. 5, the video microcode 530 on video board 204 of FIG. 2 captures local video signals as raw captured YUV bitmaps and compresses the raw captured YUV bitmaps to generate a compressed video bitstream. As part of the compression processing, video microcode 530 uses the compressed video bitstream signals to generate decompressed YUV bitmaps. In order to display a representation of the local video signals in a local monitor window, the video playback software 550 on host processor 202 of FIG. 2 transmits RGB bitmaps to the monitor 106 of FIG. 1. In preferred embodiments of the present invention, conferencing system 100 provides different options for providing video signals from the video board 204 to the host processor 202 for local monitoring.
Under option #1, only the compressed video bitstream signals are transmitted from the video board 204 to the host processor 202. Under this option, the video manager 516 directs (1) the video decode driver 548 to decompress the compressed video bitstream to generate decompressed YUV bitmaps and (2) the video playback driver 550 to color convert the YUV bitmaps to generate RGB bitmaps for display on the monitor 106.
Under option #2, the decompressed YUV bitmaps generated on the video board 204 are transmitted to the host processor with the compressed video bitstream. Under this option, the video manager 516 directs the video playback driver 550 to color convert the YUV bitmaps to generate RGB bitmaps for display on the monitor 106.
Under option #3, the decompressed YUV bitmaps are color converted on the video board 204 to generate RGB bitmaps. These RGB bitmaps are then transmitted from the video board 204 to the host processor 202 with the compressed video bitstream. Under this option, the video manager 516 directs the video playback driver 550 simply to display the RGB bitmaps on the monitor 106.
Under option #4, the raw captured YUV bitmaps are transmitted from the video board 204 to the host processor 202 with the compressed video bitstream. Under this option, the video manager 516 directs the video playback driver 550 to color convert the YUV bitmaps to generate RGB bitmaps for display on the monitor 106.
Under option #5, the raw captured YUV bitmaps are color converted on the video board 204 to generate RGB bitmaps. These RGB bitmaps are then transmitted from the video board 204 to the host processor 202 with the compressed video bitstream. Under this option, the video manager 516 directs the video playback driver 550 simply to display the RGB bitmaps on the monitor 106.
In addition to these video format options, video conferencing system 100 also preferably provides video resolution options. Under the video resolution options, when bitmaps (either YUV or RGB) are transmitted from the video board 204 to the host processor 202, the video conferencing system 100 provides the ability to transmit the bitmaps at at least two different resolutions. For example, under video format option #2, the decompressed YUV bitmaps could be transmitted at either a (160×120) resolution or a (320×240) resolution.
Those skilled in the art will understand that these video format and video resolution options provide the ability to optimize the video conferencing performance of video conferencing system 100. The different options differ from one another in three significant areas: (1) processing load to the host processor 202, (2) transmission bandwidth between the video board 204 and the host processor 202, and (3) quality of the displayed image.
Removing the decode processing from the host processor 202 reduces the host processing load, but increases the transmission bandwidth. In addition, removing the color conversion processing from the host processor 202 further reduces the host processing load, but also increases the transmission bandwidth further still (since the color converted RGB bitmaps are preferably full-resolution bitmaps while the YUV bitmaps are preferably subsampled YUV4:1:1 bitmaps). Using the raw captured video signals provides greater quality of the local video display than using the decompressed video signals. On the other hand, using the raw captured video signals implies different levels of quality between the local monitor video display and the display of the video signals received from the remote conferencing system and decoded by the host processor 202. In addition, increasing the video resolution not only increases the quality of the local monitor video display, but also increases (1) the transmission bandwidth between the video board 204 and the host processor 202 and (2) the processing load to the host processor 202.
Video conferencing system 100 is preferably able to select automatically particular video format and video resolution options to achieve video conferencing of a desired level of quality based on transmission bandwidth, processor performance, video resolution, and video format (YUV, RGB, etc.).
In alternative preferred embodiments, under options #3 and #5, the video board 204 generates the RGB bitmaps in 8 bits per pixel (bpp), color look-up table (CLUT) format. In these embodiments, the transmission bandwidth is less than that for the YUV4:1:1 bitmaps (options #2 and #4) which have an average of 9 bpp.
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. 08/157,694. In addition, a goal of self-calibration is to ensure that, during a conference call, the video decode and display overhead does not consume the participants entire PC, making either the data conferencing portion of his application or the video itself impractical. Even without concurrent data conferencing, it is conceivable that a video stream that consumes more than 100% of a host CPU could cause the video stream to fall behind the audio stream repeatedly and generate repeated video stream restarts. The outcome would be poor quality video, as measured by a decreased and inconsistent frame rate. It is advantageous that the user not reach this state accidentally, due to window re-sizing options that are generally available via the user interface.
In order to normalize the video decode overhead, file-based IRV and YUV frames are used. This video has been selected to provide "average" difficulty IRV decode. Because display timings are independent of content, display overhead can be measured with any of the IRV or YUV sequences. The PSVIDEO.INI configuration file is used to direct the self-calibration measurement and to store results. There are .INI sections for the following values:
(1) File names of the files containing the IRV and YUV video frames. These are AVI files. A FOURCC code is input to determine specify a file.
(2) The name of the SYSTEM.INI boot description display driver. This string is compared to the SYSTEM.INI entry. If a mis-compare results, then it is assumed the display driver has been changed, and self-calibration is re-run. Self-calibration should also be re-run following a processor upgrade. e.g., an OverDrive upgrade. However, self-calibration will not detect this, and the user must re-run self-calibration manually, in this case. There is a Control Panel applet for this purpose. A processor upgrade is considered a fairly rare event.
(3) Decompress calibration timings. Values for IRV local (Looking Glass), IRV remote and YUV (160×120) and (320×240) (snapshot format) are written following the calibration measurement, and later referenced in order to construct VCost function output.
(4) Display calibration timings. Values for iconic (36×36), normal (160×120), and quarter (320×240) resolution are written following the calibration measurement, and later referenced in order to construct VCost function output.
Measurement
The proc videoMeasure in module VCOST is used to perform video decode and display measurements. Results are saved in the PSVIDEO.INI entries described above. There following delineates the different measurement cases and describes the measurement steps for each case:
Iconic IRV Decode. This is not a unique case. The Normal IRV decode timing is used.
Normal IRV Decode. This is the case for remotely encoded video received over the network and displayed at normal resolution. Decode measurement is implemented by disabling Looking Glass on a local stream, causing full IRV decode to take place. An IRV AVI file is used to simulate network frames.
Quarter IRV Decode. This is the case for remotely encoded video received over the network and displayed at quarter resolution. Decode measurement is implemented by using the Looking Glass disabled local stream (previous case) and resizing the monitor window to quarter size in order to trigger codec doubling as part of the decode. An IRV AVI file is used to simulate network frames.
Iconic IRV Decode. This is not a unique case. The Normal Looking Glass timing is used.
Normal Looking Glass Decode. This is the case for local video received in RGB format from the capture hardware. Decode measurement can not actually be made, because there is no decode/color conversion in this case. Decode measurement is implemented by using a Looking Glass enabled local stream. Looking Glass can be enabled/disabled at a local stream dynamically. Because internally, normal Looking Glass frames are copied from a capture buffer to an output display buffer at monitor playback time, a copy buffer timing is used as the normal Looking Glass decode timing. An IRV AVI file is used, but the decode timing is ignored in place of an in-line (160×120×8) byte buffer copy, which simulates normal looking glass operation in this case.
Quarter Looking Glass Decode. This is the case for local video received in YUV format from the capture hardware. Decode measurement is implemented using the Looking Glass enabled local stream (previous case). Decode measurement is implemented by measuring the YUV-to-RGB color conversion step. The monitor window is resized to quarter size in order to trigger codec doubling as part of the color conversion. A YUV9 (160×120) file is used to simulate capture hardware YUV Looking Glass frames.
Because Looking Glass frames are transferred from the capture hardware to host memory in an uncompressed format (RGB or YUV9), they are relatively large in size (either 160×120×8 or 160×120×9/8 bytes=approximately 20 KBytes). Assuming a video board-to-host ISA transfer rate of 2 MBytes/sec, then Looking Glass frames will take approximately 10 msec/frame. This value is added to IRV Looking Glass decode timings to account for the system utilization impact of the Looking Glass frame transfers. In addition, so that the calibration process itself will not fully utilize the host processor, a decreased frame rate of 5 fps is used to calibrate IRV (local and remote).
Normal YUV-9 Snapshot Decode. Measures decode overhead for YUV9 (160×120) capture resolution. A YUV-9 (160×120) file is used to simulate capture hardware YUV frames.
Quarter YUV-9 Snapshot Decode. Measures decode overhead for YUV9 (320×240) capture resolution. A YUV-9 (320×240) file is used to simulate capture hardware YUV frames.
Because YUV-9 frames are transferred from the capture hardware to host memory in an uncompressed format, they are relatively large in size (either 160×120×9/8 bytes=approximately 20 KBytes or 320×240×9/8=approximately 80 KBytes). Assuming a video board-to-host ISA transfer rate of 2 MBytes/sec, then YW-9 frames will take approximately 10 msec/frame (160×120) or 40 msec/frame (320×240). This value is added to YUV-9 decode (color convert) timings to account for the system utilization impact of the YUV-9 frame transfers. In addition, so that the calibration process itself will not fully utilize the host processor, a decreased frame rate of 3 fps is used to calibrate YUV-9 snapshots. Also, because YUV-9 color convert is independent of video content (motion) and is simply a function of frame size, the same frame is repeatedly used to get an average timing.
Iconic Display. The monitor window is sized to (36×36) prior to monitor function invocation.
Normal Display. The monitor window is sized to (160×120) prior to monitor function invocation.
Quarter Display. The monitor window is sized to (320×240) prior to monitor function invocation.
When decode and display timings are derived, they are compared against values that are too much larger than the times expected on the slowest Intel® i486™ processor. For example, given that an IRV decode costs 12-24 msecs on a Intel® i486DX2™ 66 MHz processor, IRV decode times are compared against an upperbound of greater than 150 msecs. If any timings are found to be out-of-bounds, then it is assumed either another application or interrupt routine (e.g., network stack) has interfered with the calibration. In this case, an error code is returned to video manager initialization, which informs the user interface to issue a message indicating calibration should be re-run with the PC in a "quiescent" state.
File-Based Capture (File Playback)
The VInfo structure, used to specify stream attributes when a stream is opened, contains the field wCaptureSource, which can be used to specify indirectly a file-based capture for a local stream. If wCaptureSource specifies CAPT -- SOURCE -- AVI -- FILE (as opposed to CAPT -- SOURCE -- DRIVER), then the dwCaptureDriver FOURCC code is used to specify a file of AVI frames in the PSVIDEO.INI file. CAPT -- SOURCE -- DRIVER is the default so if the value in wCaptureSource is not valid, CAPT-SOURCE -- DRIVER will be assumed.
In the processing of the WM -- VCAPTURE -- CALL message to the capture task, the VINFO structure is obtained for the current stream and the wCaptureSource field is checked to determine whether to do real time video capture of AVI file based playback. This is the point, WM -- VCAPTURE -- CALL, where the flow of control diverges for the capture and file playback. If capture is specified. EnableCapture is called to start the capture driver, otherwise EnableAVIFilePlayback is called to open the file for playback.
After capture has begun or the file has been opened, the capture is driven by capture callbacks, and the playback is driven by a timer message, (ID -- AVI -- FILE -- PLAYBACK) which is processed by the play task's window procedure. EnableAVIFilePlayback executes the following:
(1) Open the correct AVI file and ascend it to the first video frame (perform file/type validation in the process).
(2) Allocate frame buffers for the capture callback.
(3) Start a timer with ID, ID -- AVI -- FILE -- PLAYBACK, and VCapt's window handle. When the WM -- TIMER message arrives, and the timer ID is equal ID -- AVI -- FILE -- PLAYBACK, the proc ReadAFrame is used to get the next video frame. The AVI "auto repeat" control is used to read a file repeatedly that consists of a limited number of frames (e.g., 1). ReadAFrame does its own frame processing without calling the capture callback function. This separation keeps messages from being sent to the driver when it was not open. The VCapt procs DisableCapture and DisableAVIFilePlayback are called from the WM -- VCAPTURE -- CALL code to stop capture and file playback, respectively.
Plavback Statistics
Once the capture source and display windows are sized appropriately for a specific calibration and VMonitor is called to begin the decode/display of the video stream, videoMeasure executes the following sequence:
SendMessage is used to send the VPlay app hwnd a WM -- MEASURE -- BEGIN message. A handle to the stream being measured is passed as the wParam.
The videoMeasure thread is delayed, using a PeekMessage loop for a duration specified (function of the stream being measured).
SendMessage is used to send the VPlay app hwnd a WM -- MEASURE -- END message. A handle to the stream being measured is passed as the wParam, and a long pointer to a measurement structure is passed as the lParam. VPlay will fill-in the measurement structure, consisting of decode and display timings for the duration of the BEGIN/END interval, in response to the WM -- MEASURE -- END message.
The VPlay statistics support measures decode and display overhead by summing the total decode/display times and then dividing the sums by the total number of frames processed. Because the statistics do not assume a frame rate or total number of frames, the videoMeasure begin/end sequence returns correct values.
VCost Function
As part of VInit, video manager initialization, the PSVIDEO.INI file is read for the measured calibration values and placed in a global data structure. When VCost is called, an internal proc videoCost is used to calculate system utilization. int videoCost.

______________________________________( IN WORD wResolution, // iconic, normal, quarter IN WORD wDispFreq, // 1=every frame, 2=every other, etc. IN WORD wFrameRate, // 10, 15, etc. IN WORD wFormat, // IRV or YUV IN DWORD dwFlags) // specifies local or remote stream______________________________________

videoCost first selects the correct decode (decompcost) and display (dispcost) costs, based on the wResolution, wFormat, and dwFlags parameters. It then calculates system utilization as follows:
decompcost=decompcost * wFrameRate;
dispcost=(dispcost * wFrameRate)/wDispFreq;
costinmsecs=decompcost+dispcost;
// Return %
return(costinmsecs/10):
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. 08/157,694.
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. 08/157,694.
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. 08/157,694.
Palette Creation
The IRV palette is preferably established if conferencing application 502 is the active application (or one of its child video windows is active) and the IRV palette is preferably mapped into the system palette, if other windows have higher Z-order. In a preferred embodiment, IRV palette creation includes the following steps:
(1) Get the IRV color table from the IRV Codec.
(2) Create a Microsoft® Windows™ operating system palette from the IRV colors.
(3) Select the palette into the display window's DC.
(4) Set the display window to the active window (when VPlay is executing, its window is active, not the display window, which was supplied via the UI/VCI).
(5) Realize the palette.
Steps (4) and (5) ensure that the foreground indices (mapping used when the window is active) are set correctly.
In an alternative embodiment, the following steps are used:
(1) Create a palette with all palette entry flags set to PC -- RESERVED. The PC -- RESERVED flag indicates that the palette entry is "animated"; i.e., the system entry (actually the hardware CLUT entry) may be changed underneath Microsoft® Windows™ palette management. With the PC -- RESERVED flag in a system palette entry, Microsoft® Windows™ operating system will not use it to map a logical palette entry. Creating a (new) palette with all palette entry flags set to PC -- COLLAPSE is not sufficient (PC -- COLLAPSE indicates a palette entry should not be mapped through an existing system palette entry), although this is what might be expected.
In fact, palette realization will begin at the first free system palette entry, if the system palette has free palette entries. Then, if more than the number of free entries are needed, system palette changes continue (wrap-around) with the first non-static entry (10th). Microsoft® Windows™ palette management does this in order to allow applications that require less than 236 system palette entries to share the system palette without re-mapping indices with every RealizePalette. However, IRV uses a full 236 color palette and IRV decode produces DIBs with a color table equal to the IRV color table. If its foreground indices have been wrapped (K-to-235, 10- K-1!, where K is the first free system palette entry), then its color display will not be correct.
(2) Set the display window to the active window.
(3) Realize the palette created above. This will set the system palette to a palette with all entries marked PC -- RESERVED. A subsequent RealizePalette (Step 4), with an active window, will begin with the 10th (first non-static) entry, and logical palette entries will be identity mapped.
(4) Create and realize the IRV palette.
After this adjustment VPlay has:
(1) Established a complete and unmodified IRV color table as the logical palette for the display window.
(2) Established identity foreground palette indices for the display window.
Both are desired for correct palettized color displays.
Extra RealizePalette Logic
A palette-aware application preferably performs the following actions in response to palette messages:

__________________________________________________________________________case WM.sub.-- PALETTECHANGED:if ((HWND) wParam == hWnd) return (0);}case WM.sub.-- QUERYNEWPALETTE:if (lpVidObj->hPalette == NULL){ DbgMsg ("vplay-vplay", VERR, "no palette to realize."); break;}SelectPalette (lpVidQbj->hDC, lpVidObj->hPalette, FALSE);changes = RealizePalette (lpVidObj->hDC);if (message==WM.sub.-- PALETTECHANGED) { DbgMsg("vplay-VidhookProc(WM.sub.-- PALETTECHANGED)", VDBG, "NUMBER PALETTE ENTRIES CHANGED = %d", changes);} else { DbgMsg("vplay-VidhookProc(WM.sub.-- QUERYNEWPALETTE)", VDBG, "NUMBER PALETTE ENTRIES CHANGED = %d", changes);}// Do realize for MSVIDEO DrawDib, too.DrawDibSetPalette(lpVidObj->hDDC, lpVidObj->hPalette);changes = DrawDibRealize(lpVidObj->hDDC, lpVidObj->hDC, FALSE);if (message==WM.sub.-- PALETTECHANGED) { DbgMsg("vplay-VidhookProc(WM.sub.-- PALETTECHANGED)", VDBG, "NUMBER PALETTE ENTRIES CHANGED for DrawDibDraw = %d", changes);} else { DbgMsg("vplay-VidhookProc(WM.sub.-- QUERYNEWPALETTE)", VDBG, "NUMBER PALETTE ENTRIES CHANGED for DrawDibDraw = %d", changes);}__________________________________________________________________________

The use of RealizePalette here is enough for the program; i.e., it is not used at the actual point of a display (blt) because either the program has been realized in the foreground, or it has been realized in some Z-order and has received all the system entries it requires or has its logical palette mapped into pre-existing system palette entries (using the Microsoft® Windows™-defined best fit algorithm).
In practice, this may not be the case. Exceptions are:
(1) A non-active RealizePalette does not always take effect. However, if a RealizePalette is made immediately before display, then a display to a window that does not have focus will be made with a correct logical-to-system mapping. Thus, although Microsoft® Windows™ palette management should not require it, RealizePalette is called immediately before every display bit. An unacceptable performance penalty has not been measured in this regard.
(2) There is a peculiarity in Microsoft® Windows™ palette managment: If a desktop bitmap is used that requires non-static palette entries, and an icon is dragged across the desktop, then the desktop is realized in the foreground and without a WM -- PALETTECHANGED message broadcast. Thus, RealizePalette is used immediately before each display bit (same as preceding case) to account for this case.
In addition to the normal palette calls, the DrawDibSetPalette and DrawDibRealize are used in parallel with SelectPalette and RealizePalette, to cause (msvideo.dll's) DrawDibDraw to use the same color palette.
There is another case where palette logic preferably has support. In the proc that has hooked the display window, if a palette changed message is received, it is first determined if the display window is iconified. Because if the display window icon is moved across a desktop bitmap, an infinite loop of RealizePalette calls may result. That is, the iconified window receives a palette changed message from the desktop, which causes a realize for the iconified window, which causes the desktop to receive a palette changed message again. Because the desktop has temporary precedence in the Z-order, an infinite loop can result. This case is eliminated with the following logic, which precedes the typical palette management logic:

__________________________________________________________________________if (message==WM.sub.-- PALETTECHANGEDmessage==WM.sub.-- QUERYNEWPALETTE)hFirstParentWnd = hParentWnd = GetParent(hWnd);if (hParentWnd){do { hOldestParentWnd = hParentWnd;} while (hParentWnd = GetParent(hParentWnd));} else {hOldestParentWnd = hFirstParentWnd;}if (hOldestParentWnd |= hFirstParentWnd) {if (IsIconic(hOldestParentWnd)) { DbgMsg("vplay-vidwndhookproc", VDBG, "RETURN -- ICONIC PARENT"); return (CallWindowProc (lpVidObj->OldRookProc, hWnd, message, wPararm, lParam));}}}__________________________________________________________________________

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. 08/157,694. In addition, the SUPERVIDEOHEADER structure is defined as follows:
MSVideoHdr
lpData
dwBufferLength
dwBytesUsed
dwTimeCaptured
dwUser
dwFlags
dwReserved 4!
Cntl
uType, uMessage
dwPktNumber
hVStrm
CompactVideoHdr
dwPackedCntl
dwTimeCaptured
dwPktNumber
bFrameData
Only CompactVideoHdr and bFrameData are sent over the network. The CompactVideoHdr fields are defined as follows:

__________________________________________________________________________dwPackedCntlwBytesUsed 16-bit word (bits 0-15 of dwPackedCntl) containing the length of the video frame data (bFrameData . . .).bFlags The low-order byte (bits 16-23 of dwPackedCntl) of the Microsoft ® Video for Windows ™ defined VIDEOHDR flags 32-bit word. Currently, only one flag value is used to mark a video frame if it is a key or reference frame. A video key frame data packet is specified as VHDR.sub.-- KEYFRAME (=8). As leftover from the VfW capture driver, bFlags bits 0 and 1 will also be set, indicating a previously prepared header and frame capture complete state. These are not cleared by NETW and can be ignored by the receiver.nbType 8 bits of type information (bits 24-31 of dwPackedCntl). Currently supported types are: VDATA (=0×01=b00000001) Video data packet. VCNTL (=0×02=b00000010) Video control packet. If VDATA, then the message contains either a key or delta frame. If VCNTL, then the message is a control message. Control types and messages are defined by bits 2,3: VCNTL.sub.-- RESTART (=0×06=b000000110) Restart control.dwTimeCaptured This field contains the video stream capture time, as derived from the source audio stream, and is used to synchronize audio and video frames at the playback site.dwPktNumber A packet sequence number, which is a monotonically increasing, unsigned integer, beginning with a value of 0 for the first frame captured.__________________________________________________________________________

The VideoHdr is used at the capture and playback sites, but is not sent over the network. The VideoHdr provides a consistent header for the capture driver, VCapt, and VPlay. VideoHdr.lpData is established once a network frame is received, and is relative to the receiver's address space.
The VideoHdrCntl adds information to the VideoHdr to support control messages between playback and capture sites. The concatenation of VideoHdr and VideoHdrCntl is referred to as the SuperVideoHdr in the Video Manager.
The CompactVideoHdr is a compact representation of the VideoHdr and VideoHdrCntI. Assuming an average IRV 1.0 delta frame size of 1000 bytes, then a 12-byte CompactVideoHdr represents an approximate 1% overhead (versus almost 5% if the SuperVideoHdr were used straightaway).
The internal procs SuperToCompact and CompactToSuper are used to translate to/from a SuperVideoHdr to a CompactVideoHdr.
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. 08/157,694.
Video Capture 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.
Custom APIs for Video Capture 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.
Video Microcode
The video microcode 530 of FIG. 5 running on video board 204 of FIG. 2 performs video compression. The preferred video compression technique is 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 according to a preferred embodiment of the present invention. 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.
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, according to a preferred embodiment of the present invention. 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.
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.
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.
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.
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.
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.
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.
Audio Subsystem Audio/Comm 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.
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.
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 board 206 of FIG. 13, according to a preferred embodiment of the present invention. 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.
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.
(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.
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.
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.
Echo Suppression Driver
The echo suppression driver (ESP) 1512 is responsible for suppressing echoes prevalent when one or both users use open speakers (rather than headphones) as an audio output device. The purpose of echo suppression is to permit two conferencing systems 100 connected by a digital network to carry on an audio conversation utilizing a particular microphone and a plurality of loudspeaker device choices without having to resort to other measures that limit or eliminate acoustic feedback ("coupling") from loudspeaker to microphone.
Specifically, measures obviated by the ESP include:
An audio headset or similar device to eliminate acoustic coupling.
A commercial "speakerphone" attachment that would perform the stated task off the PC and would add cost and complexity to the user.
The ESP takes the form of innovations embedded in the context of art known variously as "half-duplex speakerphones" or "half-duplex hands-free telephony" or "echo suppression." The ESP does not relate to art known as "echo cancellation."
The general ideas of "half-duplex hands-free telephony" are current practice. Electronic hardware (and silicon) exist that embody these ideas. The goal of this technology is to eliminate substantially acoustic coupling from loudspeaker to microphone by arranging that substantial microphone gain is never coincident with substantial speaker power output when users are speaking.
The fundamental idea in current practice is the following: Consider an audio system consisting of a receiving channel connected to a loudspeaker and a transmitting channel connected to a microphone. If both channels are always allowed to conduct sound energy freely from microphone to network and from network to loudspeaker, acoustic coupling can result in which the sound emanating from the loudspeaker is received by the microphone and thus transmitted back to the remote station which produced the original sound. This "echo" effect is annoying to users at best and at worst makes conversation between the two stations impossible. In order to eliminate this effect, it is preferable to place an attenuation device on each audio channel and dynamically control the amount of attenuation that these devices apply by a central logic circuit. This circuit senses when the remote microphone is receiving speech and when the local microphone is receiving speech. When neither channel is carrying speech energy, the logic permits both attenuators to pass audio energy, thus letting both stations receive a certain level of ambient noise from the opposite station. When a user speaks, the logic configures the attenuators such that the microphone energy passes through to the network and the network audio which would otherwise go to the speaker is attenuated (this is the "talk state"). When on the other hand speech is being received from the network and the local microphone is not receiving on speech, the logic configures the attenuators conversely, such that the network speech is played by the speaker and the microphone's acoustic energy is muted by the attenuator on that channel (this is the "listen state"). If both the local user and the remote user speak at the same time, the logic allows the speaker who is talking the loudest over their threshold to speak.
The ESP operates without a separate dedicated speakerphone circuit device. The ESP operates over a network featuring an audio codec that is permitted to distort signal energies without affecting the performance of the algorithm. The ESP effectively distributes computational overhead such that redundant signal processing is eliminated.
The ESP is a distributed digital signal processing algorithm. In the following, the algorithm is spoken of as "distributed." meaning that two instantiations of it reside on the two conferencing systems connected by a digital network, and their operation is interdependent). "Frame energy" means a mean sum of the squares of the digitized audio samples within a particular time segment called a "frame."
The instantaneous configuration of the two attenuations is encoded as a single integer variable, and the attenuations are implemented as a fractional multiplier as a computational function of the variable.
In order to classify a signal as speech, the algorithm utilizes a frame energy threshold which is computed as an offset from the mathematical mean of a histogram in which each histogram bin represents the count of frames in a particular energy range. The mean is used only if the frame is classified as noise based on the standard deviation of the histogram. The threshold varies dynamically over time as it is recalculated. There exists a threshold for each of the two audio channels.
Since both stations need access to the threshold established at a particular station (in that one station's transmit stream becomes the other station's receive stream), the threshold is shared to both instantiations of the algorithm as an out-of-band network signal. This obviates the need for both stations to analyze the same signal, and makes the stations immune to any losses or distortion caused by the audio codec.
The energy of a transmitted audio frame is embedded within a field of the communication format which carries the digitally-compressed form of the frame. In this way, the interactive performance of the station pair is immune from any energy distortion or losses involved in the audio codec.
The ESP makes possible hands-free operation for video teleconferencing products. It is well-known that hands-free audio conversation is a much more natural conferencing usage model than that of an audio headset. The user is freed from a mechanical attachment to the PC and can participate as one would at a conference table rather than a telephone call.
Audio Task Interface with 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.
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, according to a preferred embodiment of the present invention. 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.
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.
NULLDEV 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.
PWave Subsystem
The PWave subsystem provides high-priority playback of digital audio signals contained in Microsoft® standard Wave files.
PWave API
PWave API 552 of FIG. 5 is a Microsoft® Windows™ operating system DLL that accepts calls (PWave API calls) and translates them into messages sent to an onboard task (high priority playback task 554 of FIG. 5) to playback audio. This API gives the caller the ability to select a Wave file for playback at a higher priority other Wave files. By using the API, the application has control of the number of times the file is played back and when playback stops playing. The application gets notification of playback complete.
High Priority Playback Task
The high-priority playback task executes in the environment of audio/comm board 206 and is distinct from the host processor 202 and operating system where the PWave API 552 is called. This task preempts other Wave files from being played back, sets up the board for playback (board setup includes volume control and output device selection) of its Wave file, and plays back the file until it is halted or the specified number of playback loops has been reached.
PWave Protocol
All strings used in defining the PWave protocol below are NUL (`/0`) terminated ASCI strings. The following mnemonics are used in the protocol description below. Their integer values are defined as follows.

______________________________________Mnemonic Translation______________________________________POPEN.sub.-- TMSG 100PLAYDONE.sub.-- TMSG 101TMB.sub.-- EXIT-1P.sub.-- SUCCESS 100PERR.sub.-- BADFILE 101PERR.sub.-- NOTWAVEFILE 103PERR.sub.-- UNSUPPORTED.sub.-- RATE 104PERR.sub.-- NOTMONO 105PERR.sub.-- DEVBUSY 106PERR.sub.-- NOTRIFF 107PERR.sub.-- UNRECOGIZABLE.sub.-- WAVE 108PERR.sub.-- UNSUPPORTED.sub.-- SAMPLE 109PERR.sub.-- RESOURCE.sub.-- FAILURE 110PERR.sub.-- FILEACCESS 111______________________________________

The PWave task can be opened in two ways: (1) to test a Wave file for compatibility with the implementation and (2) to playback a file. Method (1) takes a single parameter on the command line of the high-priority task. That parameter is a string specifying the full path name to the file. The message passed back from the board is of the following format:

__________________________________________________________________________MESSAGE Parameter Meaning__________________________________________________________________________POPEN.sub.-- TMSG P.sub.-- SUCCESS File is valid.POPEN.sub.-- TMSG PERR.sub.-- BADFILE Unexpected file data.POPEN.sub.-- TMSG PERR.sub.-- NOTMONO Invalid audio type.POPEN.sub.-- TMSG PERR.sub.-- NOTWAVEFILE Invalid file type.POPEN.sub.-- TMSG PERR.sub.-- UNRECOGIZABLE.sub.-- WAVE Bad format type.POPEN.sub.-- TMSG PERR.sub.-- NOTRIFF Bad header type.POPEN.sub.-- TMSG PERR.sub.-- RESOURCE.sub.-- FAILURE On board resource error.POPEN.sub.-- TMSG PERR.sub.-- UNSUPPORTED.sub.-- RATE Sample too slow/fast.POPEN.sub.-- TMSG PERR.sub.-- UNSUPPORTED.sub.-- SAMPLE Wrong number of bits per sample.POPEN.sub.-- TMSG PERR.sub.-- DEVBUSY High priority audio channel in use.POPEN.sub.-- TMSG PERR.sub.-- FILEACCESS Can't open file.__________________________________________________________________________

Playing back the file requires five parameters on the high priority task's command line. All parameters are strings passed to the task, and converted and interpreted in that environment. The indexes are as follows:
(1) The file name. This is a full path to the Wave file to be played by the "High Priority Playback" task. This path is a NUL terminated ASCII string.
(2) Number of times to loop on the file. If the number to loop is non-zero, the file is played that many times. If the number to loop is specified as zero, the file is played indefinitely (until the caller stops it). The number is a string specified with any number of the following digit concatenated together: 0, 1, 2, 3, 4, 5, 6, 7, 8, 9. It is converted to a 32-bit unsigned integer.
(3) Ignored parameter.
(4) The volume to playback the Wave file. This is a string composed of one or more of the following digits: 0, 1, 2, 3, 4, 5, 6, 7, 8, 9, and at most one ".". It is converted to a 32-bit floating point number that the audio samples are multiplied by before being played.
(5) Output device specification. An integer string that is any of the following numbers added together 1, 2, 4 as integers and converted to a string. The integers have the following specifications:
1 specifies the on-board speaker.
2 specifies the line-out device.
4 specifies the head phone device.
If the call is accepted, the following message will be sent to the host:

______________________________________MESSAGE Parameter Meaning______________________________________POPEN.sub.-- TMSG P.sub.-- SUCCESS File is valid.______________________________________

Error conditions returned to the host are as follows:

__________________________________________________________________________MESSAGE Parameter Meaning__________________________________________________________________________POPEN.sub.-- TMSG PERR.sub.-- FILEACCESS Can't open file.POPEN.sub.-- TMSG PERR.sub.-- BADFILE Unexpected file data.POPEN.sub.-- TMSG PERR.sub.-- NOTMONO Invalid audio type.POPEN.sub.-- TMSG PERR.sub.-- NOTWAVEFILE Invalid file type.POPEN.sub.-- TMSG PERR.sub.-- UNRECOGIZABLE.sub.-- WAVE Bad format type.POPEN.sub.-- TMSG PERR.sub.-- NOTRIFF Bad header type.POPEN.sub.-- TMSG PERR.sub.-- RESOURCE.sub.-- FAILURE On board resource error.POPEN.sub.-- TMSG PERR.sub.-- UNSUPPORTED.sub.-- RATE Sample too slow/fast.POPEN.sub.-- TMSG PERR.sub.-- UNSUPPORTED.sub.-- SAMPLE Wrong number of bits per sample.POPEN.sub.-- TMSG PERR.sub.-- DEVBUSY High priority audio channel in__________________________________________________________________________ use.

To stop audio playback, the following message is passed from the host to the board via the boards device driver:

______________________________________MESSAGE Parameter Meaning______________________________________TMB.sub.-- EXIT None Stop execution.______________________________________

At the completion of playing this file for any reason (i.e., application request or number of times played), the following message is sent back to the host:

______________________________________MESSAGE Parameter Meaning______________________________________PLAYDONE.sub.-- TMSG P.sub.-- SUCCESS File is done. Task cleanup complete.______________________________________

Comm Subsystem
The communications (comm) subsystem of conferencing system 100 of FIG. 5 comprises comm API 510, comm manager 518, and DSP interface 528 running on host processor 202 of FIG. 2 and comm task 540 running on audio/comm 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, according to a preferred embodiment of the present invention. The comm subsystem consists of the following layers that reside both on host processor 202 and the audio/comm board 206:
Transport independent interface (TII.DLL),
Datalink module (DLM.DLL+KPDAPI.DLL, where KPDAPI.DLL is the back-end of the DLM which communicates with the DSP interface), and
Reliable datalink module (RDLM.DLL).
TII.DLL and RDLM.DLL reside entirely on the host processor. Datalink module comprises DLM.DLL residing on the host processor, and control (D channel), D channel driver, data comm tasks, and B channel drivers residing on audio/comm board 206.
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. In a preferred embodiment, 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). Alternative preferred embodiment may use alternative transport media such as local area networks (LANs) as the communication network.
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, according to a preferred embodiment of the present invention. The comm subsystem provides an asynchronous interface between the audio/comm 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 a 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 ISDN audio/comm board. The comm subsystem is responsible for providing the ISDN B channel device drivers for the ISDN audio/comm board. The comm subsystem provides the ISDN D channel device drivers for the ISDN audio/comm board. The comm software is preferably certifiable in North America (U.S.A., Canada) and Europe. The signaling software is compatible with NI1, AT&T Custom, and Northern Telecom DMS-100.
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.
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.
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.
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. 08/157,694. Comm API 510 supports calls to three different types of transport-independent interface functions by conferencing applications 502 and 504 to the comm subsystem: connection management functions, data exchange functions, session management, and communications statistics functions. Connection management functions provide the ULM with the ability to establish and manage virtual channels for its peers on the network. Data exchange functions control the exchange of data between conferencing systems over the network. Communications statistics functions provide information about the channels (e.g., reliability, priority, number of errors, number of receives and transmissions). These functions are as follows:

__________________________________________________________________________Connection Management FunctionsRegisterChanMgr Registers a callback or an application window whose message processing function will handle low-level notifications generated by data channel initialization operations. This function is invoked before any OpenChannel calls are made.RegisterChanHandler Registers a callback or an application window whose message processing function will handle low-level notifications generated by data channel input/output (I/O) activities. The channels that are opened will receive CHAN.sub.-- DATA.sub.-- SENT, and the accepted channels will receive CHAN.sub.-- RECV.sub.-- COMPLTE.OpenChannel Requests a sub-channel connection from the peer application. The result of the action is given to the application by invoking the callback routine specified in the RegisterChanHandler. The application specifies an ID for this transaction. This ID is passed to the callback routine or posted in a message. Note: All Connection requests are for establishing connections for sending data. The receive channels are opened as the result of accepting a ConnectChannel request.AcceptChannel A peer application can issue AcceptChannel in response to a CHAN.sub.-- REQUEST (OpenChannel) message that has been received. The result of the AcceptChannel call is a one-way communication sub- channel for receiving data. Incoming data notification will be sent to the callback or window application (via PostMessage) to the ChannelHandler.RejectChannel Rejects an OpenChannel request (CHAN.sub.-- REQUEST message) from the peer.CloseChannel Closes a sub-channel that was opened by AcceptChannel or ConnectChannel.Data Exchange FunctionsSendData Sends data. Data is normally sent via this mechanism.ReceiveData Receives data (normally received through this mechanism). Call is normally issued in response to a DATA.sub.-- AVAILABLE message.Communications Statistics FunctionsGetChanInfo Returns channel information.GetChanStats Returns various statistical information about a channel.GetTiiStats Returns various statistical information about a TII__________________________________________________________________________ channel.

These functions are defined in further detail later in this specification in APPENDIX E entitled "Comm API."
In addition, comm API 510 supports three types of messages and callback parameters returned to conferencing applications 502 and 504 from the comm subsystem in response to some of the above-listed functions: session messages, connection messages, and channel messages. Session messages are generated in response to change of state in the session. Connection messages are generated in response to the various connection-related functions.
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.
Session Handler Messages
The description for this section is the same as the description for the section of the sane name in U.S. patent application Ser. No. 08/157,694. One difference is in the CONN -- ACCEPTED message, which is defined as follows:

__________________________________________________________________________CONN.sub.-- ACCEPTED Response to MakeConnection or AcceptConnection request.wParam Connection handlelParam Pointer to connection information structure: { DWORD TransId (specified by user in earlier request) LPCONN.sub.-- CHR Pointer to connection attributes LPBYTE Itone string }__________________________________________________________________________

Channel Manager Messages
The description for t his section is the same as the description for the section of the same name in U.S. patent application Ser. No. 08/157,694.
Channel Handier 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.
iTone
The iTone version string is a C (null-terminated) string that contains version information for the product. This string is in the first packet sent over the iTone channel after it is opened. In a preferred embodiment, the iTone message is transmitted and received over a channel (e.g., channel 15) different from the control channel to avoid interference with control channel messages. The iTone message is sent even before the control channel is opened. Since iTone is the first packet placed on the wire, this ensures that no other messages are sent until iTone transmission is complete. To guarantee delivery of the iTone message, an acknowledgment method is employed in which the iTone packet itself is used for acknowledgement.
The generic format for the string is:
"iTone rev company.product.version copyrightyear"
where:

______________________________________∘ company company ID (5 characters, zero-filled)∘ product company product ID (4 characters, zero-filled)∘ version product version ID (4 characters, zero-filled)∘ year copyright year______________________________________

The quotes are not part of the string. All spaces in the string are single spaces, and the company, product, and version IDs are separated by a period. Preferred company IDs are as follows:

______________________________________Company iTone Company Number______________________________________Intel 00001VTel 00002______________________________________

Preferred product IDs are as follows:

______________________________________Company iTone Company Number______________________________________Intel Video Conferencing (XXXXXXXX) 0001Intel Data Conferencing (XXXXX) 0002VTel Video Conferencing 0003______________________________________

Preferred company.product.version IDs are as follows:

______________________________________Company Product Shipping Version iTone Number______________________________________Intel VC 1.0 00001.0001.0000Intel VC 1.6 00001.0001.0006Intel VC 1.8 00001.0001.0008Intel VC 2.0 00001.0001.0020______________________________________

For example, the preferred iTone string for version 1.6 of the Intel Video Conferencing product is as follows:
"iTone rev 00001.0001.0006 copyright1994"
The iTone acknowledgment scheme has three goals:
1. Guarantee that the iTone message is delivered.
2. Accommodate delays in non-end-to-end digital calls, but do not penalize end-to-end digital callers with a delay.
3. Use the iTone packet itself as the acknowledgment medium, to guarantee synchronization between both participants in the call.
To use the iTone packet for acknowledgment, one octet within the stream is used as a state flag. (Note that this octet is not a part of the iTone version string.) This state flag has three possible values. For the sake of this description, these values and the packets in which they reside are called iTone1, iTone2, and iTone3. The following is a description of a typical scenario of iTone transmissions for a preferred acknowledgment method:
1. With caller and callee in iTone1 state, caller transmits first iTone stream (iTone1) to callee. (Callee may also be transmitting iTone1 to caller at the same time.)
2. Callee receives iTone1 from caller and callee switches to iTone2 state.
3. Callee transmits second iTone packet (iTone2) to caller.
4. Caller receives iTone2 from callee and caller switches to iTone2 state.
5. Caller transmits third iTone packet (iTone3) to callee and caller switches to iTone3 state.
6. Callee receives iTone3 from caller and callee switches to iTone3 state.
Once iTone acknowledgment completes, TII saves the iTone version pointer string and proceeds with the rest of connection establishment. If iTone acknowledgment does not complete within a specified time period (e.g., 20 seconds), the session establishment fails and the connection between VC systems is not completed.
During the iTone Channel Open callback, a buffer large enough to receive the iTone packet is posted. TII then begins transmitting iTone1 packets at preferably 200 millisecond intervals. During iTone1 and iTone2 processing, packets are sent at preferably 200 millisecond intervals.
The ConAcc structure within the SESS -- CB structure contains a long pointer field (called "lpItone"). This structure is used during TII's callback to VCI. When the session is established and the CONN -- ACCEPT callback is sent to VCI, the pointer to the version string is included in this structure.
The SESS -- CB structure with the IpItone member is as follows:

______________________________________// lParam structure for Session handler// (in cases where multiple parameters are returned via lParam)typedef struct tagSESS.sub.-- CB {union tagSESS.sub.-- EV {struct tagConReq {HSESS hSess;LPTADDR lpCallerAddr;LPCONN.sub.-- CHR lpAttributes;} ConReq;struct tagConAcc {DWORD dwTransId;LPCONN.sub.-- CHR IpAttributes;LPBYTE lpItone;} ConAcc;} SESS.sub.-- EV;} SESS.sub.-- CB, FAR *LPSESS.sub.-- CB;______________________________________

When VCI receives the CONN -- ACCEPT callback from TII, the pointer to the iTone version string (lpItone) is saved. The V -- ConAcc structure within the LPV -- CBACK structure contains a long pointer field (lpItone). This structure is used during VCI's callback to the application. When the session is established and the CFM -- ACCEPT -- NTFY callback is sent to the application, the pointer to the version string is included in this structure. The LPV -- CBACK structure with the lpItone member is as follows:

______________________________________// lParam structure for Callbacks// (in cases where multiple parameters are returned via lParam)typedef struct tagV.sub.-- CBACK {union tagV.sub.-- EV {struct tagV.sub.-- ConReq {LPTADDR lpCallerAddr;LPCONN.sub.-- CHR lpAttributes;} V.sub.-- ConReq;struct tagV.sub.-- ConAcc {LPCONN.sub.-- CHR lpAttributes;LPABBUSCARDINFO lpABCard;LPMTYPE lpMedia;LPBYTE lpItone;} V.sub.-- ConAcc;struct tagV.sub.-- ChanReq {DWORD dwTransId;HCONN hCall;LPCHAN.sub.-- INFO lpChanInfo;} V.sub.-- ChanReq;} V.sub.-- EV;} V.sub.-- CBACK, FAR *LPV.sub.-- CBACK;______________________________________

When the application receives its CFM -- ACCEPT -- NTFY callback, it preferably looks at the new lpItone pointer within the LPV -- CBACK structure it receives. lpItone is the address of a buffer that contains the iTone version string. The buffer may be parsed to extract the iTone version information. Parsing preferably follows the syntax for iTone version strings as described earlier in this section. If the contents of the buffer do not follow this exact syntax, the version string is assumed to be invalid.
The following code defines the iTone sub frame and the buffers used for sending and receiving iTone. In addition, the bit positions of the three major iTone components are defined. These components are the version string and the iTone state flag, as well as the bit rate flag.

__________________________________________________________________________// iTone data pattern for a sub.sub.-- multiframe//#define FAS.sub.-- 0 0×00#define FAS.sub.-- 1 0×01#define BAS.sub.-- 0 0×00static BYTE ItoneSubFrame ! =FAS.sub.-- 0, FAS.sub.-- 0, FAS.sub.-- 0, FAS.sub.-- 1, FAS.sub.-- 1,FAS.sub.-- 0, FAS.sub.-- 1, FAS.sub.-- 1, BAS.sub.-- 0, BAS.sub.-- 0,BAS.sub.-- 0,BAS.sub.-- 0, BAS.sub.-- 0, BAS.sub.-- 0, BAS.sub.-- 0, BAS.sub.-- 0,`i`, `T`, `o`, `n`, `e`, ` `, `r`, `e`, `v`, ` `, `0`, `0`,`0`, `0`, `1`, `.`, `0`, `0`, `0`, `1`, `.`, `0`, `0`, `0`, `6`, ` `,`c`, `o`, `p`, `y`, `r`, `i`, `g`, `h`, `t`,`1`, `9`, `9`, `4`, NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL,NULL,NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL,NULL, NULL, NULL, NULL, FAS.sub.-- 0, FAS.sub.-- 1, FAS.sub.-- 0,FAS.sub.-- 1, FAS.sub.-- 1, FAS.sub.-- 0, FAS.sub.-- 1,FAS.sub.-- 1, BAS.sub.-- 0, BAS.sub.-- 0, BAS.sub.-- 0, BAS.sub.-- 0,BAS.sub.-- 0, BAS.sub.-- 0, BAS.sub.-- 0, BAS.sub.-- 0, `i`, `T`, `o`,`n`,`e`, ` `, `r`, `e`, `v`, ` `, `0`, `0`, `0`, `1`, `.`, `0`, `0`, `0`,`1`, `.`, `0`, `0`, `0`, `6`, ` `, `c`,`o`, `p`, `y`, `r`, `i`, `g`, `h`, `t`, `1`, `9`, `9`, `4`, NULL, NULL,NULL, NULL, NULL,NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL,NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL};//// Position of "iTone" string inside first sub-multiframe//#define ITONE.sub.-- ID.sub.-- POS 16//// Position of state flag inside first sub-multiframe//#define ITONE.sub.-- STATE.sub.-- POS 56//// Position of bit rate flag inside first sub-multiframe//#define ITONE.sub.-- BITRATE.sub.-- POS 57//// Size of each iTone sub-multiframe//#define ITONE.sub.-- SUB.sub.-- FRAMESIZE sizeof(ItoneSubFrame)#define NUM.sub.-- SUBFRAMES 5//// Size of HDLC hardware-imposed trailer (2 byte flag + 2 byte checksum)//#define ITONE.sub.-- HW.sub.-- TRAILER.sub.-- SIZE 4//// Size of combined iTone message for both ISDN B channels.// This will be split into two fragments, one for each B channel.// Each fragment will be preceded by the HDLC header and QMUX header.// Each fragment will be trailed by the HDLC trailer (flag+checksum).//#define ITONE.sub.-- MSGSIZE (ITONE.sub.-- SUB.sub.-- FRAMESIZE*NUM.sub.-- SUBFRAMES - ITONE.sub.-- HW.sub.-- TRAILER.sub.-- SIZE) * 2//// Buffer used to fill and send iTone//static BYTE ItoneBuf ITONE.sub.-- MSGSIZE+ITONE.sub.-- HW.sub.-- TRAILER.sub.-- SIZE*2!;//// Buffer used to receive iTone//static BYTE ItoneRecvBuf ITONE.sub.-- MSGSIZE + ITONE.sub.-- HW.sub.--TRAILER.sub.-- SIZE*2!;//// iTone Stream State (per connection)//#define ITONE.sub.-- STREAM.sub.-- 1 0×01#define ITONE.sub.-- STREAM.sub.-- 2 0×02#define ITONE.sub.-- STREAM.sub.-- 3 0×03//// Rate Negotiation State (per connection)//#define RATE.sub.-- STREAM.sub.-- 1 0×02#define RATE.sub.-- STREAM.sub.-- 2 0×03#define RATE.sub.-- STREAM.sub.-- 3 0×04#define RATE.sub.-- ESTAB 0×ff__________________________________________________________________________

iTone as implemented by an iTone data handler (H.221) on the board is a simple step of transmitting iTonel until it receives iTone1 (or times out). If the iTone handler (H.221) detects iTone1, it abandons iTone1 transmission/reception and just informs the host that iTone was detected. Subsequently, TII starts the 3 phases of iTone all over again in QMUX mode. If any of these assumptions is not correct, H.221 will not be able to talk to R1.6.
The iTone packet does not have 2 HDLC octets preceding the "iTone.." ASCII string; so the ASCII string is transmitted and received right after the 16 FAS/BAS octets. A single HDLC header and QMUX header is transmitted preceding the 6 iTone frames. (1 HDLC flag octet, plus 8 QMUX header octets). The above QMUX header ensures the receipt of iTone. There is an arbitrary number of HDLC flags in between these "super" iTone messages (each of which has 6 consecutively packed iTone frames).
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.
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), according to a preffered embodiment of the present invention. 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, according to a preferred embodiment of the present invention. 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, according to a preferred embodiment of the present invention. 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, according to a preferred embodiment of the present invention. 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, according to a preferred embodiment of the present invention. 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 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.
Data Link Manager
Referring now to FIG. 29, there are shown diagrams indicating typical 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. 08/157,694. One difference is the addition of the DLM -- ChangeBitRate function, which is defined as follows:

______________________________________WORD DLLEXP DLM.sub.-- ChangeBitRate (DWORD DlmConnId, // Connection handleDWORD CallReference, // Connection Call referenceDWORD ReceiveBitRate, // BITRATE.sub.-- 112KB or BITRATE.sub.-- 128KBDWORD TransmitBitRate) // BITRATE.sub.-- 112KB or BITRATE.sub.-- 128KB______________________________________

The DLM -- ChangeBitRate function directs the low level comm software to change bit rate on the B channel for receiving and transmission of data.
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.
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.
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.
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, according to a preferred embodiment of the present invention. Source video is video that is captured (and optionally monitored) on the local conferencing system and sent to the comm subsystem for transmission to a remote system. Sink video is video that is captured remotely, received from the comm subsystem, and played back on the local system. The video packet fields are defined as follows:

__________________________________________________________________________Length 16-bit word containing the length of the video frame data in octets.Flag Used to mark a video frame if it is a key (i.e., reference) frame. When key frames are transmitted in slices, each slice has this field set to indicate a key frame. A key frame is indicated by: VHDR.sub.-- KEYFRAME(=8): Video key frame data packet. A delta frame is indicated by: VHDR.sub.-- DELTAFRAME(=0): Video delta frame data packet.Reserved 4 bits reserved for future expansion.Type Defines the type of the frame. Preferably set to: VDATA(=1): Video data packet.TimeCaptured Set to the time, in milliseconds, between the beginning of the capture session and the current frame. In a preferred embodiment, this field is used to carry a time stainp used to synchronize audio and video frames at the receiving node.PktNumber A packet sequence number, which is a monotonically increasing, unsigned integer, beginning with a value of 0 for the first frame captured in the capture session.FrameData The compressed video data which may represent either a key frame or a delta frame. This field is only present if the packet type is__________________________________________________________________________ VDATA.

A video data frame containing compressed video generated with the Intel® Indeo C™ compression algorithm preferably does not exceed 8192 octets in length. For this reason, key frames at the (320×240) resolution may be transmitted in slices.
A conferencing node may request a key frame from a remote note by sending the Video Request Key Frame control message through the video control channel (i.e., a virtual circuit). This control message is defined as follows:

______________________________________Length (Bits) Field______________________________________28 bits Reserved4 bits Type8 octets Reserved______________________________________

Type defines the type of the frame and is preferably set to:
VCNTL (=2): Video control packet.
The Reserved fields are all reserved for future extension and are preferably all zero.
Compressed Video Bitstream
Referring now to FIG. 25, there is shown a representation of the compressed video bitstream for conferencing system 100, according to a preferred embodiment of the present invention. 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.
Video Decoding Procedure
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.
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.
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.
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.
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.
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.
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.
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.
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.
Audio Protocol
Referring now to FIG. 26, there is shown a representation of a compressed audio packet for conferencing system 100, according to a preferred embodiment of the present invention. Source audio is audio that is captured (and optionally monitored) at the local system and sent to the comm subsystem for transmission. Sink audio is audio that is received from the comm subsystem for playback on the local system. Audio is preferably handled on audio/comm board 206 and not on host processor 202. The compressed audio packet of FIG. 26 is that which is actually sent/received from the communications subsystem and not necessarily that manipulated by an application on the host processor. The audio packet fields are defined as follows:

__________________________________________________________________________Timestamp 30-bit value used to synchronize audio and video frames at the receive endpoint. The audio stream preferably generates timestamps as a master clock that are copied to the captured video frames before transmission.Mute Bit indicates whether or not the audio stream is muted or not. The audio is muted when the bit is set. When the Mute bit is set, no audio data is sent.Extension Extension bit (preferably 0).Data Compressed audio data.__________________________________________________________________________

The length of the audio data is not explicitly specified in the packet header. A receiving endpoint's comm subsystem reassembles an audio packet and therefore implicitly knows the length and can report it to its application. The length of an audio packet is a run-time parameter and depends on the compression method and the amount of latency desired in the system. The preferred audio compression/decompression method implementation has 100-millisecond latency, which translates to 180 bytes of compressed audio data per packet.
Compressed Audio Bitstream
The preferred audio stream for conferencing system 100 is a modification of the European Groupe Speciale Mobile (GSM). GSM was developed in the context of the standardization of the European digital mobile radio. It resulted from the combination of the Regular-Pulse Excitation/Linear-Predictive-Coding codec developed by Philips (Germany) with the Multi-Pulse-Excitation/Linear-Predictive-Coding codec devised by IBM (France). For further information, see the ETSI-GSM Technical Specification. GSM 06.10, version 3.2.0, UDC 621.396.21, published by the European Telecommunication Standards Institute in Valbonne Cedex, France.
The data rate of the standard GSM codec is 13.0 kbits/sec. The preferred GSM implementation for conferencing system 100 has a bit rate of 16 kbits/sec. The mean opinion score (MOS) quality rating of the preferred GSM implementation is 3.54. It is not prone to rapid quality degradation in the presence of noise. The relative complexity is about 2 MOPSs/s. Due to implementation processing considerations, the standard GSM implementation is adjusted to yield the preferred GSM implementation. In addition, headers are added to provide extra control information, such as frame counting and muting.
In order to save processing, the 260-bit audio frame is not packed. This results in a 320-bit frames. These frames occur every 20 milliseconds. This increases the bit rate from 13 kbits/sec to 16 kbits/sec. The composition of the preferred audio frame is as follows:

______________________________________typedef struct {unsigned int lar1: 6; /* stp parameters */ unsigned int lar2: 6; unsigned int lar3: 5; unsigned int lar4: 5; unsigned int lar5: 4; unsigned int lar6: 4; unsigned int lar7: 3; unsigned int lar8: 3; } STP;typedef struct {unsigned int lag 7; unsigned int gain 2; /* ltp parameters */ unsigned int grid 2; /* rpe parameters */ unsigned int xmax 6; unsigned int x0 3; /* pulse amplitude*/ unsigned int x1 3; unsigned int x2 3; unsigned int x3 3; unsigned int x4 3; unsigned int x5 3; unsigned int x6 3; unsigned int x7 3; unsigned int x8 3; unsigned int x9 3; unsigned int x10 3; unsigned int x11 3; unsigned int x12 3; } LTP.sub.-- RPEtypedef struct {STP frame; LTP.sub.-- RPE subframe(4); } GSMBITS;______________________________________

The result of not packing these structures on a Texas Instrument® C31 DSP, a 32-bit processor, is a 320-bit frame. At a frame rate of 50 frames/sec, the data rate is 16.0 kbits/sec.
A header has also been added to groups of frames. The length of the header is one 32-bit word. The MSB is an extension bit reserved for future extension. Bit 30 is a mute flag (1=mute). The remaining bits represent a timestamp. This time stamp is not actually time, but is preferably a frame counter. The initial value of it is arbitrary. It is therefore a relative number representing the progress of audio frames and useable for synchronization.
Data Protocol
Data packets are inside TII packets. The data conferencing application will have its own protocol inside the TII protocol stack. Data conferencing application 504 is described in greater detail in U.S. patent application Ser. No. 08/137,319 (filed Oct 14, 1993) and in U.S. patent application Ser. No. 08/170,146 (filed Dec. 20, 1993).
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.
Reliable Transport Comm Protocols
Referring now to FIG. 27, there is shown a representation of the reliable transport comm packet structure, according to a preferred embodiment of the present invention. 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.
Unreliable Transport Comm Protocols
At the lowest layer of conferencing system 100, an unreliable protocol is preferably used to transport data on the ISDN B-channels. For those applications requiring reliability, the reliable protocol discussed in the previous section is added on top of the unreliable protocol discussed in this section. The unreliable protocol sits atop of HDLC framing which the unreliable protocol uses for actual node-to-node transport of packets. Even though HDLC framing is used, a data link protocol is not implemented. In particular, there is no guarantee that data packets will be delivered or that they will be uncorrupted at the receive node of a link. The CRC validation of the HDLC is used to detect corrupted data.
The unreliable protocol provides for logical channels and virtualization of the two Basic Rate ISDN B-channels. Logical channels are local site entities that are defined between the DLM and TII is layer and the client (i.e., application program) using them. The logical channels provide the primary mechanism clients use to send multiple data types (e.g., audio, video, data). The layer services multiplex these data types together for transmission to the remote sites.
In a preferred embodiment, logical channel zero is used as a control channel. Site peers (i.e., two conferencing systems in a conferencing session) use this control channel to exchange information on their use of other logical channels. Logical channels are half-duplex. Therefore, two channels are necessary to send and receive data. A priority attribute is associated with a logical channel (and therefore with a data type). The unreliable protocol asserts that higher priority data will always be sent ahead of lower priority data when both are pending. Priorities are assigned by an API call to the TII services. Audio has the highest priority, then data, and last video.
Although the ISDN Basic Rate Interface (BRI) defines two physical 64 kbit/second B-channels for data, the services at both DLM and TII virtualize the separate B-channels as a single 128 kbit/second channel. Client data types, defined by their logical channels, are multiplexed into a single virtual stream on this channel. In a preferred embodiment, this inverse multiplexing is accomplished by breaking all packets into an even number of fragments and alternating transmission on the two physical B-channel connections. Initially, after channel establishment, the first fragment is sent on the B1-channel, the second on the B2-channel, etc. At the receiving site, fragments are collected for reassembly of the packet.
Referring now to FIG. 28, there is shown a representation of the unreliable transport comm packet structure, according to a preferred embodiment of the present invention. The fields of the preferred unreliable transport comm packet are defined as follows:

__________________________________________________________________________Flag Standard HDLC Flag field.iLINK Link layer header specific to network. Most significant bit (MSB) = 1; next most significant bit (NMSB) = 0. The low octet of the 16-bit iLINK field is equal to the QMUX sequence number.iSEQ Multilink layer for joining muitiple links (e.g., channels B1 and B2 in an ISDN call). MSB = 1; NMSB = 0. The low octet of the 16-bit iSEQ field is null.iMULTI Multipoint addressing and routing of data streams. If iSEQ = QMUX, then the high octet of the 16-bit iMULTI field the receiver logical channel number (LC#) and the low octet is the sender LC#.iSTX Virtual circuit addressing. If iSEQ = QMUX, then the high octet of the 16- bit iSTX field = start of message (SoM) / end of message (EoM) and the low octet is the message number (lower six bits).Data The data field.CRC Standard HDLC CRC field.Flag Standard HDLC Flag field.__________________________________________________________________________

Feature and Capability Negotiation
Conferencing Management
This section describes the messages and protocols for conferencing management functions. During the initiation of a conference call, the conference manager 544 of FIG. 5 establishes a connection with its peer at the remote conference node utilizing the underlying communication mechanism (shown as TII 510 in FIG. 5). The conference manager 544 will then open a pair of reliable control channels (one incoming and one outgoing channel) with its peer to carry out conferencing management functions. These conferencing management function include conference login, capability negotiation, unreliable audio/video channels and subsystems setup, reliable video key frame control channels setup, and participant information exchange.
Connection Management
Connection and Channel Setup
The conference manager 544 makes an ISDN call to its peer on a remote conference node. The conference manager 544 then sets up two one-way reliable control channels (one outgoing and one incoming channel) by accessing the communication services provided by the communication subsystem. The following steps show the sequence of how the connection and different channels are set up in a typical conference.
(1) Caller makes an ISDN call to the remote conference node.
(2) Callee rejects or accepts the call. If callee rejects the call, then processing terminates; otherwise, processing continues to next step.
(3) Caller opens a reliable control channel of type CFI -- CTRL -- TYPE (=0).
(4) Callee accepts the control channel request.
(5) Callee opens a reliable control channel of type CFI -- CTRL -- TYPE (=0).
(6) Caller accepts the control channel request.
(7) Conference login is implemented.
(8) Capability negotiation is started.
(9) Capability negotiation is completed with agreement as to audio, video, and data conferencing capabilities.
(10) Caller opens audio channel of type CFI -- AUD -- TYPE (=1).
(11) Callee accepts audio channel request.
(12) Caller opens video channel of type CFI -- VID -- TYPE (=2).
(13) Callee accepts video channel request.
(14) Caller opens reliable video key frame control channel of type CFI -- VIDKEY -- TYPE (=3).
(15) Callee accepts video key frame channel request.
(16) Callee opens an audio channel of type CFI -- AUD -- TYPE (=1).
(17) Caller accepts audio channel request.
(18) Callee opens video channel of type CFI -- VID -- TYPE (=2).
(19) Caller accepts video channel request.
(20) Callee opens reliable video key frame control channel of type CFI -- VIDKEY -- TYPE (=3).
(21) Caller accepts video key frame channel request.
(22) Participant information exchange is implemented.
Connection Shutdown and Error Handling
The connection between two conference nodes can be shutdown using the normal call hang-up process at any time. The typical process to shutdown the connection involves:
Terminating the conference (initiated by either conference node).
Leaving the conference (initiated by the conference node that leaves the conference).
Errors (initiated by either or both conference nodes of the connection, when an unrecoverable error is encountered at any time).
The correct sequence to shut down a connection is to free up all local resources, close all channels, and then hang up the call. However, every conference node is expected to clean up local resources and recover from errors whenever a call hang-up request is received from the remote node or initiated by the local node. Among errors that would result in a hang-up is a time-out imposed by the conferencing application on the period required to complete connection and channel setup.
Conference Login
After the physical call connection and logical reliable control channels are established, the conference manager 544 of the caller will start the conference login process (step #7 above). Conference login, which is implemented after the connection and control channels have been established, involves the following steps:
(a) Caller sends a login request message, identifying the caller node type and (optionally) the caller user name.
(b) If the login request is accepted, then callee sends a positive login response, identifying the callee node type and (optionally) the callee user name. If the login request is rejected, then callee sends a negative login response, identifying the reason for the rejection.
After successful conference login, the conference nodes are ready for capability negotiation. The format for the login request message is as follows:

______________________________________Field Data Type Description______________________________________Message Number 16-bit integer 100Node Type 16-bit integer 2Name Length 16-bit integer Length of name (0 => no name is provided)Name 256 characters User name (maximum length)Reserved 40 characters Reserved for future extension______________________________________

The format for the login positive response message is as follows:

______________________________________Field Data Type Description______________________________________Message Number 16-bit integer 101Node Type 16-bit integer 2Name Length 16-bit integer Length of name (0 => no name is provided)Name 256 characters User name (maximum length)______________________________________

The format for the login negative response message is as follows:

______________________________________Field Data Type Description______________________________________Message Number 16-bit integer 102Reason 16-bit integer 0 => unspecified 1 => unauthorized user ID 2 => unsupported node type______________________________________

Capability Negotiation
Capability negotiation (step #8 above) allows two conference nodes to agree upon a set of basic capabilities and subsystem protocols to be used during a conference. Negotiation is based upon a simple message protocol initiated by the caller and terminated by the callee. Messages contain a Request/Response field followed by a Capabilities structure.
Capabilities Structure
The Capabilities structure begins with a length and personal conferencing specification (PCS) version ID fields. The remaining fields in the structure are defined by the PCS version ID. In PCS 1.0, the Capabilities structure is as follows:

______________________________________Field Data Type Description______________________________________ILen 32-bit integer Length of this structurePCS Version 32-bit integer PCS version ID; 1 => Version 1.0Product 32 characters Vendor/product nameVersion 6 characters Product version (e.g., V1.6, V2.0)Audio Caps 32-bit integer Bit-wise-OR of supported audio capabilities 0 × 1 => no audio 0 × 2 => PCS GSM v1.0Video Caps 32-bit integer Bit-wise-OR of supported video capabilities 0 × 1 => no video 0 × 2 => Intel ® Indeo C ™ v2.0______________________________________

The audio and video capability fields are bit masks that can be set by a conferencing node to advertise all protocols and capabilities supported by the node. Alternatively, a node can set one bit in each field thereby defining the specific protocols the node wants to use during the conference. The Capabilities structure is defined to contain a unique set if only one bit is set in each of the capability fields. Bits in the AudioCaps and VideoCaps fields represent bit stream protocols supported by the node's audio and video subsystems, respectively. For example, AudioCaps bit 0×2 maps to the audio bit stream GSM protocol defined in Appendix B of PCS. VideoCaps bit 0×2 represents the Intel® Indeo C™ v2.0 bit stream defined in Appendix C of PCS with (160×120) base resolution, a maximum of 10 frames per second, and 84 kbps data rate.
Requests and Responses
The Request/Response field that prefixes the Capabilities structure is a 16-bit integer. The following requests and responses are preferably supported in capability negotiation.
CallerCapRequest
The caller sends the CallerCapRequest message to begin capability negotiation. The caller's Capabilities structure is set up to identify either (1) all capabilities and protocols supported by the conferencing node or (2) the unique set of capabilities the node wants to use for the conference. If the structure is unique, the caller expects the callee to accept or reject the unique configuration via the CalleeCapAccept or CalleeCapReject responses, respectively. Otherwise, when the set is not unique, the caller expects the callee to return a subset (possibly unique) of these capabilities via CalleeCapResponse. The CallerCapRequest/CalleeCapResponse process continues until a unique set of capabilities is chosen by a node. The negotiation terminates when the callee acknowledges this set via the CalleeCapAccept response. The format for the CallerCapRequest message is as follows:

______________________________________Field Data Type Description______________________________________Message Number 16-bit integer 500Capabilities Structure See Capabilities structure defined above.______________________________________

CallerCapCancel
If an error in the protocol is detected, the caller sends the CallerCapCancel message to terminate the negotiation The callee does not respond to this request. The caller can send this message in response to a CalleeCapAccept message received with unexpected capabilities. The format for the CallerCapCancel message is as follows:

______________________________________Field Data Type Description______________________________________Message Number 16-bit integer 501Capabilities Structure See Capabilities structure defined above.______________________________________

CalleeCapAccept
The callee sends the CalleeCapAccept message to accept the caller's proposed or negotiated unique conference capability configuration thereby terminating the capability negotiation process. The format for the CalleeCapAccept message is as follows:

______________________________________Field Data Type Description______________________________________Message Number 16-bit integer 502Capabilities Structure See Capabilities structure defined above.______________________________________

CalleeCapReject
The callee sends the CalleeCapReject message to reject the caller's unique conference capability configuration. This terminates the capabilities negotiation process. The callee returns in the message its capabilities thereby allowing the caller to note incompatibilities. In addition, the callee can use this message to terminate the negotiation if an error in the protocol is detected.

______________________________________Field Data Type Description______________________________________Message Number 16-bit integer 503Capabilities Structure See Capabilities structure defined above.______________________________________

CalleeCapResponse
The callee sends the CalleeCapResponse to return to the caller a subset derived from among the alternative capabilities presented by the previous CallerCapRequest message. If a unique set is returned, the caller is expected to acknowledge this selection with a CallerCapRequest message. Otherwise, the request/response process will repeat until a unique set of capabilities is chosen by a node. When a unique set is finally received from the caller, the CalleeCapAccept message will be returned by the callee thereby terminating the negotiation.

______________________________________Field Data Type Description______________________________________Message Number 16-bit integer 504Capabilities Structure See Capabilities structure defined above.______________________________________

The following list of steps illustrates the use of these messages in a first possible scenario for capabilities negotiation:
(1) Caller begins negotiation by sending a CallerCapRequest message with a unique set of capabilities.
(2) Callee responds by sending either (1) a CalleeCapAccept message with an acknowledgment of the accepted unique set of capabilities or (2) a CalleeCapReject message identifying the callee's capabilities. In either case, negotiation is now complete.
The following list of steps illustrates the use of these messages in a second possible scenario for capabilities negotiation:
(1) Caller begins negotiation by sending a CallerCapRequest message with a non-unique set of capabilities.
(2) Callee responds by sending a CalleeCapReject message identifying the callee's capabilities. Negotiation is now complete.
The following list of steps illustrates the use of these messages in a third possible scenario for capabilities negotiation:
(1) Caller begins negotiation by sending a CallerCapRequest message with a non-unique set of capabilities.
(2) Callee responds by sending a CalleeCapResponse message identifying a subset of the caller's capabilities.
(3) Caller sends a CallerCapRequest message identifying a subset of the capabilities identified by callee in step (2).
(4) Callee responds by sending a CalleeCapResponse message identifying a subset of the capabilities identified by caller in step (3).
(5) Caller sends a CallerCapRequest message identifying a unique set of capabilities which is also a subset of the capabilities identified by callee in step (4).
(6) Callee responds by sending a CalleeCapAccept message with an acknowledgment of the accepted unique set of capabilities. Negotiation is now complete.
Inclusion of the product name and version fields in the Capabilities structure allows for two nodes to recognize each other as peers after the first request and response. In such cases, the nodes could optionally continue negotiation using their own custom protocol. Adding support for new audio or video protocols or data capabilities is as simple as defining new bits in the AudioCaps and VideoCaps fields. Extending the bit definitions does not require a revision of the protocol.
Participant Information Exchange
Once a conference has been established, the caller sends a Conference Participation Information (CPI) message to the callee. Upon receiving the caller's CPI, the callee responds by sending its own CPI back to the caller. This message exchange occurs across the TII control channel. The structure of the CPI message is as follows:

______________________________________Field Data Type Description______________________________________User Id 16-bit Unique user Id in a conferenceTitle 10 chars Mr., Ms., Mrs., Dr., etc.Last Name 64 characters User's last nameFirst Name 64 characters User's first nameJob Position 64 characters e.g., President, Manager, EngineerCompany Name 64 characters User's company nameStreet 80 characters Street nameCity 32 characters City nameState/Province 32 characters State or province namePostal Code 32 characters Zip code or postal codeCountry 32 characters Country nameEmail 64 characters Email addressTelephone #1 32 characters Office phone numberTelephone #2 32 characters Office phone numberFax # 32 characters Fax number______________________________________

Conference Participation Messages
The caller sends the ConferenceAnnouncePresence message to the callee to announce the presence of the conference. The callee responds by returning the same message containing its own CPI. The User ID field in the CPI structure is undefined in the point-to-point conference case. The structure of the ConferenceAnnouncePresence message is as follows:

______________________________________Field Data Type Description______________________________________Message Number 16-bit integer 600CPI Structure See CPI structure defined above.______________________________________

Flow Control Over Reliable Channels
When conferencing signals are transmitted over reliable channels from a conferencing system A to a conferencing system B of FIG. 1 during a conferencing session, conferencing system B transmits an acknowledgment message to notify conferencing system A of the receipt of each signal packet. The acknowledgment message informs conferencing system A that the signal packet it previously sent was received by conferencing system B. Conferencing system A may also use the acknowledgment message as authorization to send additional signal packets to conferencing system B.
If conferencing system B sends acknowledgment messages as soon as it receives signal packets, but before it has had a chance to process those signal packets, then conferencing system B runs the risk of being inundated with subsequent signal packets that it may not be able to handle. If conferencing system B drops one or more signal packets, then conferencing system A will have to retransmit those signal packets. To help alleviate this problem, in a preferred embodiment of the present invention, conferencing system B does not send acknowledgment messages until it knows it can handle the subsequent signal packets that may be sent by conferencing system A in response to the acknowledgment messages.
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, according to a preferred embodiment of the present invention. As shown in FIG. 36, conferencing system A transmits signal packets to conferencing system B over the reliable channels (step 3602) and conferencing system B subsequently receives those signal packets from conferencing system A (step 3604). Rather than sending the acknowledgment messages for the received signal packets right away, conferencing system B begins to process the received signal packets (step 3606). Conferencing system B sends the acknowledgment messages to conferencing system A as the received signal packets are processed (step 3608). Conferencing system A receives the acknowledgment messages from conferencing system B (step 3610) and uses those acknowledgment messages as authorization to send additional signal packets and repeat the processing of steps 3602 to 3610.
By delaying the transmission of the acknowledgment messages until the signal packets are processed, conferencing system B reduces the chances that it will not be able to handle the additional signal packets that may be sent by conferencing system A in response to the acknowledgment messages.
Preemptive Priority-Based Transmission
Conferencing system 100 of FIG. 1 is able to transmit different types of signals during a conferencing session with a remote conferencing system. For example, a conferencing session using both audio/video conferencing application 502 and data conferencing application 504 involves the transmission of audio signals, video signals, and data signals. In addition, the following different types of control signals may be transmitted:
TII control A signals--used by TII 510 for rate negotiation.
TII control B signals--used by TII 510 to establish other channels.
User interface control signals--used by conference manager 544 to exchange information about current status of conferencing applications.
VCI control signals--used by VCI 506 to exchange identification information about users.
Video control signals--used to ask for key frames.
Although these different signals are transmitted through different virtual channels, they are all preferably transmitted over the same physical connection (e.g., the same IDSN line).
In order to provide high quality audio conferencing, it is important to have a steady flow of audio signals between the conferencing systems. If the flow of audio signals is not steady enough, then the playback of audio signals may be interrupted causing the quality of the audio playback to be adversely affected. The audio subsystem of conferencing system 100 is designed to generate packets of compressed audio signals at a rate sufficient to provide high quality audio playback.
Since audio channels share the same connection with video channels (as well as channels for other types of signals), special handling is preferably performed to avoid the unnecessary interruption of the flow of audio signals over the connection. If signals were transmitted in the order in which they were generated, then the quality of at least the audio portion of the conferencing session may be adversely affected. For example, conferencing system 100 generates audio packets more frequently than video packets and the video packets are typically larger than the audio packets. If the transmission of a relatively large video packet were allowed to complete before any subsequently generated audio packets were transmitted, then the delay in transmitting the audio packets may adversely affect the quality of the audio playback.
To reduce the chances of such adverse affects to conference quality, in a preferred embodiment of the present invention, each different type of signal is assigned a priority level and the order of transmitting those signals is based on the relative priority levels.
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, according to a preferred embodiment of the present invention. FIG. 37 demonstrates the transmission processing for two types of signals, where the first type of signals has a higher priority than the second type of signals. The processing shown in FIG. 37 begins when the comm subsystem receives one or more packets of the second type of signals (step 3702). If those packets are too large to be transmitted over the connection intact, then the comm subsystem breaks the packets into smaller fragments for transmission and begins to transmit the fragments over the connection to the remote conferencing system (step 3704).
In the processing of FIG. 37, it is assumed that the comm subsystem receives one or more packets of the first type of signals before the transmission of all of the fragments of the second type of signals is complete (step 3706). In response, since the priority of the first type of signals is higher than the priority of the second type of signals, the comm subsystem interrupts the transmission of the packets of the second type of signals, breaks the packets of the first type of signals into fragments (if necessary), and transmits all of the packets of the first type of signals (step 3708). After the transmission of the packets of the first type of signals is complete, the comm subsystem resumes the transmission of the packets of the second type of signals (step 3710).
In a preferred embodiment, where a packet of the second type of signals is broken into two or more fragments, the comm system may interrupt the transmission of the packet between fragments. For example, if a video packet is broken into 8 fragments and an audio packet becomes available for transmission after 3 of the 8 video fragments have been transmitted, then the comm subsystem preferably transmits all of the fragments of the audio packet before resuming the transmission of the video fragments with the fourth video fragment.
Those skilled in the art will understand that the processing of FIG. 37 can be extended to three or more types of signals were the different types of signals have different levels of priority. In a preferred embodiment, the different types of signals have the following priority levels assigned to them, where higher priority levels imply higher priority for transmission:
Audio signals: Priority level 15
TII control A signals: Priority level 14
TI control B signals: Priority level 14
User interface control signals: Priority level 14
VCI control signals: Priority level 13
Data signals: Priority level 12
Video control signals: Priority level 12
Video signals: Priority level 10
If two different types of signals have the same priority level, then the relative order for transmission of those signals is the order in which they are generated.
It will also be understood that the preemptive priority-based transmission of signals according to the present invention eliminates the need for bandwidth reservation for different types of signals. It will be further understood that the worst-case latency for conference system 100 may be controlled by adjusting the size of the fragments into which the signal packets are broken.
Rate Negotiation
Some telecommunications switches may deliver a 64 kilobit per second (KBPS) data call as 56 KBPS without notifying the calling node. The calling node assumes that a clear 64 KBPS channel connection has been made, but the called node has a 56 KBPS connection. This data rate mismatch prevents any HDLC data exchange on the B channels.
According to a preferred embodiment of the present invention, both nodes (i.e., the calling node and the called node) initially configure their B channels at 56 KBPS. This is done even if the switch presents a 64 KBPS clear channel to the calling node. The product identification packet is exchanged between the two nodes at 56 KBPS. One octet within the product identification packet indicates the call rate. For the calling node, this octet indicates the rate at which the call was placed. For the called node, the octet is the rate at which the call was presented. By the time the exchange of product identification packets is complete, both nodes know what each other considers to be the call rate.
If either party believes that the call rate is 56 KBPS, then no rate negotiation takes place. Since the product identification packets are exchanged at 56 KBPS, and since one or both nodes believes that the call rate is 56 KBPS, then no rate negotiation is needed. The remainder of the call remains at 56 KBPS.
If, however, at the end of the exchange of the product identification packets, both nodes believe that the call is at 64 KBPS, then rate negotiation begins. B channel rate negotiation occurs in-band starting at 56 KBPS. A four-way message handshake ensures that both sides may upgrade their B channels and successfully communicate at 64 KBPS.
The rate negotiation preferably begins immediately after the product identification packets have been exchanged at 56 KBPS. Negotiation messages are sent as QMUX messages on channel 15, which is the same channel used for the product identification packets. These messages are exchanged only if both sides believe that 64 KBPS data exchange is possible.
The upgrade to 64 KBPS data exchange preferably occurs via a four-way message handshake as follows:
(1) The called node upgrades its B channel receiver to 64 KBPS. Both nodes begin transmitting a stream of Rate -- Request(rate1) messages at 56 KBPS. The streams consist of messages sent at an interval of 200 milliseconds.
(2) When the calling node receives a Rate -- Request message, it upgrades its B channel transmitter and receiver to 64 KBPS. The calling node stops its stream of Rate -- Request messages at 56 KBPS and begins transmitting a Rate -- Response(rate2) message stream at 64 KBPS.
(3) Upon receiving a Rate -- Response message, the called node upgrades its B channel transmitter to 64 KBPS, stops transmitting its Rate -- Request messages at 56 KBPS, and begins transmitting a Rate -- Confirm(rate3) message stream at 64 KBPS.
(4) Upon receiving a Rate -- Confirm message, the calling node stops transmitting its Rate -- Response messages and transmits a single Rate -- Established(rate -- est) message. Upon receiving a Rate -- Established message, the called node stops transmitting its Rate -- Confirm messages.
After this four-way message handshake, rate negotiation is complete for both nodes and control channel initialization may begin. Rate -- Request messages are transmitted at 56 KBPS. Any extra transmissions after the calling party has upgraded to 64 KBPS will be ignored (e.g., as bad HDLC frames). Rate -- Response, Rate -- Confirm, and Rate -- Established messages are exchanged at 64 KBPS.
The calling node is free to transmit any message after it has sent a Rate -- Established message. However, the calling node continues to acknowledge further Rate -- Confirm messages with Rate -- Established messages.
If rate negotiation does not successfully complete within 10 seconds, both nodes will revert to 56 KBPS. Rate negotiation will restart, but the contents of the messages would indicate that the B channels be retained at 56 KBPS by both nodes. Under normal circumstances, rate negotiation will then complete successfully at 56 KBPS. If, for some reason, rate negotiation at 56 KBPS is unsuccessful (e.g., due to switch problems), then the call will be terminated with an error indicating that rate negotiation timed out.
Video conferencing systems according to a preferred embodiment of the present invention are capable of operating in a loopback mode. In loopback mode, when a called node receives an call from a calling node, the called node forwards the call back to the calling node. In that way, the calling node also becomes a called node. Rate negotiation is designed such that most of its implementation does not distinguish between calling and called nodes, since each node may have to perform as both a calling node and a called node. Most actions within the rate negotiation disregard whether the receiver is the calling or called node. The response to a given rate negotiation message usually depends on the message received. There are two exceptions to this role-independence, neither of which affect loopback operations. First, only the called node performs the initial 64 KBPS upgrade. Second, the called node takes precautions to ignore rate1. If rate1 is not ignored, then the called node may hamper negotiations during rare race conditions. If the called node actually responded to rate1 by upgrading its transmitter and receiver to 64, then it will lose communication with the calling node which is still at 56, and may not yet have seen rate1.
Referring now to FIGS. 38-41, there are shown state diagrams for the rate negotiation processing, according to a preferred embodiment of the present invention. FIG. 38 shows the state diagram for the complete rate negotiation processing. FIG. 39 shows the state diagram for the rate negotiation processing for a called node during a 64 KBPS upgrade. FIG. 40 shows the state diagram for the rate negotiation processing for a calling node during a 64 KBPS upgrade. FIG. 41 shows the state diagram for the rate negotiation processing in loopback mode during a 64 KBPS upgrade.
Certain state transitions shown in FIG. 38 are missing from FIGS. 39 and 40. This is because Rate1 is ignored by the called node, and in fact may not be received due the mismatch between rates at startup. In that case, Rate2 will not be transmitted by the called node; Rate3 will then not be transmitted by the calling node; and Rate -- Est will then not be transmitted by the called node. Unlike standard calls the loopback line of FIG. 41 goes through all the states from Rate1 through Rate -- Est.
In a preferred embodiment, TII is presented with a single virtual call that combines two physical B channel calls. TII then conducts a single rate negotiation for both B channels simultaneously. A rate upgrade to 64 KBPS occurs only if the calls on both B channels were originated and delivered as 64 KBPS. Otherwise, both B channels will be used at 56 KBPS and the extra 8 KBPS that may be available on one of the B channels is not used.
The preferred rate negotiation messages are defined by the following structure:

______________________________________typedef struct tagRATE.sub.-- REQUEST { BYTE Type; BYTE Protocolld; DWORD Rate;} RATE.sub.-- REQUEST, FAR *LPRATE.sub.-- REQUEST;______________________________________

The fields are defined as follows:
Type: the type of rate message currently being sent. During the exchange process, this type will have the following defined values:

______________________________________#define RATE.sub.-- STREAM.sub.-- 1 0 × 02#define RATE.sub.-- STREAM.sub.-- 2 0 × 03#define RATE.sub.-- STREAM.sub.-- 3 0 × 04#define RATE.sub.-- ESTAB 0 × ff______________________________________

ProtocolId: the version protocol of TII. The preferred defined value for this field is as follows:

______________________________________#define TII.sub.-- PROTOCOLID 0______________________________________

Rate: the current rate under negotiation. The defined values are as follows:

______________________________________#define BITRATE.sub.-- 112KB 0#define BITRATE.sub.-- 128KB 2______________________________________

In a preferred embodiment, rate negotiation may be disabled using ICOMM.INI. To disable negotiation, the RateHandling entry in the TII section is used. The preferred values for RateHandling are as follows:

______________________________________Flag Effect______________________________________0 Enable rate negotiation (default)1 Disable rate negotiation; use rate of call as made/presented2 Disable rate negotiation; use 56 KBPS3 Disable rate negotiation; use 64 KBPS>3 Disable rate negotiation; use rate of call as made/presented______________________________________

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.
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.