What is video conferencing?
Video conferencing is a communication session between two users or a group of users, regardless of their location, while participants see and hear each other according to the rules defined by the type of video conference.
Video conferencing is conducted subject to the use of special tools that can be implemented either on the basis of hardware solutions and systems, or in the form of software for PCs, mobile devices, or browsers.
To provide participants with sound and picture, various peripheral equipment is used: cameras, screens, microphones, speakerphones, headsets, congress systems and projectors. As a data transmission medium can be used as a network of enterprises, built on different principles, and the global Internet.
Modern video and audio codecs, specialized network protocols, various signal processing algorithms make it possible to achieve high-quality communications on virtually any communication channels.
Often, during a video conference session, a demonstration of various media data is necessary; for this, video conferencing systems allow you to capture and transmit to remote participants presentations, images of the desktop or its individual windows, as well as documents of various formats. This is achieved by using special software, additional cameras (for example, documentary cameras), capturing the signal from video outputs of laptops, PCs and other systems, including medical systems.
To summarize Video conferencing is a high-tech modern communication tool designed to improve business efficiency, streamline business processes, speed up decision making and save money on business trips.
Types of video conferencing
There are two main types of video conferencing - personal and group. Personal video conferencing involves a video session in which only two subscribers participate. By group video conferencing is meant all other types of video conferencing. Different established rules for displaying video conference participants for each of the parties are called video conferencing types. We suggest to understand this question in more detail!
Video conferencing 1-on-1
Everything is simple here: two subscribers participate, both see and hear each other at the same time. Immediately make a reservation that during any session of a video conference can use various tools for collaboration, such as the exchange of text messages, files, presentations and other media data.
Symmetrical video conferencing
They are video conferencing with a constant presence, from the English. Continuous Presence. This is the name of a video conference session in which more than 2 people participate and all participants see and hear each other at the same time. Naturally, video conferencing involves full-duplex communication . In other words, this is an analogue of a round table, where everyone has equal rights. Group video conferencing is suitable for meetings where maximum participation of each participant is required.
Voice activated video conferencing
The name of this mode is derived from the English designation Voice Activated Switching (VAS). This videoconferencing assumes the following communication format: all participants in the session hear and see only the speaker who speaks on their screens, while he sees himself or the previous speaker. Small variations of this mechanism are possible, but the essence remains the following: the VCS server tracks the voice activity of subscribers and switches the image broadcast to all participants, to the speaker. This mode has significant drawbacks, for example, false positives for noise, cough or mobile phone ring.
Selector Video Conferencing
The mode in which participants are divided into two types: speakers and students, where each participant can become a speaker (with the permission of the conference organizer). The leader of such a conference himself appoints the speakers and can remove them from the video podium at any time.
This mode may also be referred to as role video conferencing. Selector video conferencing is used most often during web conferencing (webinars).
Video conferencing for distance education
A special mode in which all participants (students) see and hear only one broadcasting user (teacher), and he sees and hears all students. Pupils are not distracted by each other, and the teacher controls them.
Video broadcast
A kind of video conference in which the speaker broadcasts to a wide audience of listeners, while he does not see or hear them. Other participants see and hear only the speaker. Feedback is possible only via text chat. Often, to smooth out changes in network conditions, during the broadcast a significant delay is introduced up to several seconds between the broadcasting and the listeners.
Video conferencing equipment
Depending on the location and method of connecting to the videoconferencing session, different peripheral equipment may be required.
Video conferencing is a communication session between two users or a group of users, regardless of their location, while participants see and hear each other according to the rules defined by the type of video conference.
Video conferencing is conducted subject to the use of special tools that can be implemented either on the basis of hardware solutions and systems, or in the form of software for PCs, mobile devices, or browsers.
To provide participants with sound and picture, various peripheral equipment is used: cameras, screens, microphones, speakerphones, headsets, congress systems and projectors. As a data transmission medium can be used as a network of enterprises, built on different principles, and the global Internet.
Modern video and audio codecs, specialized network protocols, various signal processing algorithms make it possible to achieve high-quality communications on virtually any communication channels.
Often, during a video conference session, a demonstration of various media data is necessary; for this, video conferencing systems allow you to capture and transmit to remote participants presentations, images of the desktop or its individual windows, as well as documents of various formats. This is achieved by using special software, additional cameras (for example, documentary cameras), capturing the signal from video outputs of laptops, PCs and other systems, including medical systems.
To summarize Video conferencing is a high-tech modern communication tool designed to improve business efficiency, streamline business processes, speed up decision making and save money on business trips.
Types of video conferencing
There are two main types of video conferencing - personal and group. Personal video conferencing involves a video session in which only two subscribers participate. By group video conferencing is meant all other types of video conferencing. Different established rules for displaying video conference participants for each of the parties are called video conferencing types. We suggest to understand this question in more detail!
Video conferencing 1-on-1
Everything is simple here: two subscribers participate, both see and hear each other at the same time. Immediately make a reservation that during any session of a video conference can use various tools for collaboration, such as the exchange of text messages, files, presentations and other media data.
Symmetrical video conferencing
They are video conferencing with a constant presence, from the English. Continuous Presence. This is the name of a video conference session in which more than 2 people participate and all participants see and hear each other at the same time. Naturally, video conferencing involves full-duplex communication . In other words, this is an analogue of a round table, where everyone has equal rights. Group video conferencing is suitable for meetings where maximum participation of each participant is required.
Voice activated video conferencing
The name of this mode is derived from the English designation Voice Activated Switching (VAS). This videoconferencing assumes the following communication format: all participants in the session hear and see only the speaker who speaks on their screens, while he sees himself or the previous speaker. Small variations of this mechanism are possible, but the essence remains the following: the VCS server tracks the voice activity of subscribers and switches the image broadcast to all participants, to the speaker. This mode has significant drawbacks, for example, false positives for noise, cough or mobile phone ring.
Selector Video Conferencing
The mode in which participants are divided into two types: speakers and students, where each participant can become a speaker (with the permission of the conference organizer). The leader of such a conference himself appoints the speakers and can remove them from the video podium at any time.
This mode may also be referred to as role video conferencing. Selector video conferencing is used most often during web conferencing (webinars).
Video conferencing for distance education
A special mode in which all participants (students) see and hear only one broadcasting user (teacher), and he sees and hears all students. Pupils are not distracted by each other, and the teacher controls them.
Video broadcast
A kind of video conference in which the speaker broadcasts to a wide audience of listeners, while he does not see or hear them. Other participants see and hear only the speaker. Feedback is possible only via text chat. Often, to smooth out changes in network conditions, during the broadcast a significant delay is introduced up to several seconds between the broadcasting and the listeners.
Video conferencing equipment
Depending on the location and method of connecting to the videoconferencing session, different peripheral equipment may be required.
Video conferencing in a meeting room or congress hall
In order to equip a meeting room with high quality, it is necessary to observe many nuances. Naturally, the more of them, the higher the cost of preparation. First of all, it is necessary to correctly calculate and install a sound reinforcement system; there was a good report on this topic at one of Video + Conferences . If the room is small, then it will be enough to install one or more speakerphones (these are special devices that combine one or more microphones and speakers, and are designed to eliminate echoes and noises).
Next, you need a PTZ video camera, it differs from the usual one with the ability to rotate, bend up and down, as well as zoom in and out. Such a camera can both in manual and automatic modes (for this you will need special equipment) to switch between the speakers and the audience. As a display system, the use of two LCD screens of a large diagonal is recommended: one for video participants, the second for presentations and other content.
Well, not the last place is occupied by the interior of the room: good illumination, contrasting, but not bright background on the walls, noise absorbing panels and so on. As can be seen, the cost of the equipment of the meeting room, depending on the chosen video conferencing solution, peripheral equipment and finishes, may differ by an order of magnitude.
Video conferencing in the workplace
There are already many ready-made kits and complexes, which include everything you need, but taking up extra space on the table. Therefore, often, and also in order to save, they use an ordinary working PC as a video conferencing terminal, since there is no difference in quality with the right choice of peripherals between it and specialized hardware systems.
To prepare a PC for a videoconferencing session, you need a good webcam (see the list of recommended equipment ), unfortunately, most of the integrated mono blocks and laptops cameras are not suitable for video conferencing. A headset (preferably a USB headset) or a portable speakerphone connected to a PC via a USB interface.
Mobile video conferencing
One of the advantages of video conferencing is their mobility. They can be used even while traveling or on the go. A device that can act as a video conferencing terminal - a smartphone, tablet computer, or even a clock. It is enough to install a special application on them. The manufacturers of these devices have already taken care of the rest: a front-facing camera, a powerful central processor, hardware support for video codecs (which is also needed for watching movies or YouTube), and a good speaker and microphone is a matter of course. This method of video conferencing will allow you to be always in touch with your colleagues, business partners, friends or relatives, regardless of the circumstances.
On the other hand, there are a number of difficulties associated with mobile video conferencing, some industries have yet to be resolved in order to make them truly convenient and popular, like on a PC.
What influences the quality of video conferencing?
In contrast to the usual electronic communications, such as e-mail or messaging, video conferencing refers to the so-called real-time communication (from the English. Real Time Communications), which impose more serious requirements, both on video conferencing terminals and channels ties binding them.
We all used to judge the quality of the connection by its speed, which in the context of a videoconferencing would not be entirely true. The declared speed can change rapidly in time, can decrease under load, and can drastically differ from the direction of transmission. While all of this is critical for video conferencing, where the uniformity and predictability of data flow is most important. Video conferencing is not difficult to adjust the video stream for a wide range of values from 64 KB / s to, say, 4 Mb / s, depending on the type of conference and signal quality of participants. It is much more difficult in real time to adapt the channel width to the changing conditions of each participant in the communication session.
In real conditions, the type of architecture used for organizing video conferences and the ability of this architecture to work in constantly changing conditions comes to the first place in assessing the quality of video conferencing:
- CPU power of terminal terminals. In parallel with the communication session, the user can begin to perform resource-intensive tasks.
- Video capture capabilities on the terminal camera. The camera may have excellent resolution, but give a “noisy” picture of poor quality with a lack of lighting.
- Video conference display options on the terminal screen. For example, a user has left full-screen mode and now he does not need to send video in high quality.
- The width of the channel between the video conferencing server and between participants. This is the most common situation. It can have many variations: someone in the organization began to download a large amount of data from the network and drastically reduced network resources for video conferencing. Or you, communicating via video from a smartphone, hit a crowded place, and the nearest base station of your telecom operator can no longer guarantee you the same speed and quality of the connection.
The simplest solution to this problem is hard redundancy, both hardware and network resources of the video conferencing system. But at the same time, this solution is the most expensive. Fortunately, science and technology are not standing still, and modern video conferencing systems can guarantee the excellent quality of communication in any conditions by using modern software architectures. Let's look at this issue in more detail.
Types of video conferencing system architectures
Any group videoconferencing, obviously, requires a mechanism and method for organizing data transfer between its participants. Since the transfer between the participants directly on the principle of a fully connected graph (each to each) is of little use in practice (see the actual conditions of the previous section), we will consider options using a certain medium, let's call it "VCS server", t. e. system, working on topology “star” (from the center to each).
In traditional hardware video conferencing systems, such a server is called an MCU; there is no well-established name in software. The server task is switching and processing flows during group video conferencing. The VCS server is the core of the VCS infrastructure, the resources of which are used by the VCS terminals.
Traditionally, it was decided to divide all solutions into software and hardware by type of architecture, but in 2014 such a division was not relevant for several reasons. First of all, because there are both hardware solutions that use the architecture inherent in software solutions (based on switching and SVC), as well as software solutions that copy the working principle of the MCU. And, secondly, all leading manufacturers are trying to shift their video conferencing infrastructure to virtualized environments and deliver them as software.
Mixed Videoconferencing Architecture (MCU)
During video conferencing, the video conferencing server receives video streams from each of the participants , decodes them, reduces them, stitches together the new image of the required quality and resolution for each participant (do not forget about the actual conditions described above), re-encodes and sends. All this requires tremendous computing power, introduces processing delays on the server, and can also degrade the quality as a result of data compression.
The scalability of such an architecture is extremely low, even considering the possibility of its virtualization, so the price for such an infrastructure is extremely high, and in modern realities such expenses are simply not justified.
Multiplex video conferencing architecture (Switching)
This is a classic approach to building software systems of video conferencing, according to this principle, for example, Skype works. Unlike the MCU, the VKS server in this case does not bother with trans coding, creates copies of incoming streams and sends them to other participants “as is”. It turns out that each terminal receives several video streams in full quality at once, which it simply cannot display simultaneously. The terminal has to reduce the resolution of each incoming video from each of the participants on its side, or ask to reduce it before sending, which degrades the quality of the video for all other participants.
Plus, this scheme has one: the infrastructure is not demanding of resources and even an ordinary PC can withstand hundreds of such conferences at the same time. But there are much more minuses: the terminal (usually a simple PC) has to decode not one, but several streams at once, and the video conferencing server requires several times the outgoing width of the channel to accommodate all copies of the streams created by it.
Add to this the real conditions, and get a system that hardly “digests” the number of participants more than 3, and dramatically worsens the quality of the video for everyone, when a mobile subscriber who is not able to “digest” the original picture quality sent by others subscribers.
Scalable Video Coding (SVC) based video conferencing architecture
This architecture combines all the advantages of a mixing approach and, at the same time, is devoid of the disadvantages of systems based on multiplexing. It is cheap, instantly scaled and works on all platforms. This was made possible by the development of signal processing and data compression technologies.
The bottom line is that the terminal compresses its video stream in layers: each additional layer increases the video resolution, its quality and the number of frames per second. If the channel between the terminal and the VKS server is good, then the terminal sends the maximum possible number of layers. It is worth noting that the layer is not a separate video stream of lower quality, but a full-fledged difference between it and the previous layer. Thus, the SVC stream is only 15-20% different in channel width from a non-SVC stream, and is significantly less than the required amount of bandwidth of independent streams.
The VCS server, having received an SVC stream with layers, simply cuts off the extra ones without recoding, only by discarding data packets from it according to certain rules. Thereby, allowing on the fly to create individual sets of video streams (“layouts” of windows) for each of the participants of a group video conference, depending on its actual communication conditions.
Using modern protocols and codecs
For organizing video conferencing between various software and third-party equipment, standard data transfer protocols are used.
H.239 is a communication protocol supporting two media streams from different sources. Suitable for video conferencing in which the image is displayed on two different screens (for example, in the video conference room, when on one screen - the image of the speaker, on the second - the accompanying presentation).
H.323 is a protocol for transmitting data over networks with non-guaranteed bandwidth. It is used in personal and multipoint video conferencing.
SIP is a network protocol for establishing a connection between client applications from different manufacturers, replacing the H.323 standard. Used in video conferencing and IP telephony.
Compressing and playing audio and video during a conference call is done through the use of audio and video codecs.
Types of video conferencing system architectures
Any group videoconferencing, obviously, requires a mechanism and method for organizing data transfer between its participants. Since the transfer between the participants directly on the principle of a fully connected graph (each to each) is of little use in practice (see the actual conditions of the previous section), we will consider options using a certain medium, let's call it "VCS server", t. e. system, working on topology “star” (from the center to each).
In traditional hardware video conferencing systems, such a server is called an MCU; there is no well-established name in software. The server task is switching and processing flows during group video conferencing. The VCS server is the core of the VCS infrastructure, the resources of which are used by the VCS terminals.
Traditionally, it was decided to divide all solutions into software and hardware by type of architecture, but in 2014 such a division was not relevant for several reasons. First of all, because there are both hardware solutions that use the architecture inherent in software solutions (based on switching and SVC), as well as software solutions that copy the working principle of the MCU. And, secondly, all leading manufacturers are trying to shift their video conferencing infrastructure to virtualized environments and deliver them as software.
Mixed Videoconferencing Architecture (MCU)
During video conferencing, the video conferencing server receives video streams from each of the participants , decodes them, reduces them, stitches together the new image of the required quality and resolution for each participant (do not forget about the actual conditions described above), re-encodes and sends. All this requires tremendous computing power, introduces processing delays on the server, and can also degrade the quality as a result of data compression.
The scalability of such an architecture is extremely low, even considering the possibility of its virtualization, so the price for such an infrastructure is extremely high, and in modern realities such expenses are simply not justified.
Multiplex video conferencing architecture (Switching)
This is a classic approach to building software systems of video conferencing, according to this principle, for example, Skype works. Unlike the MCU, the VKS server in this case does not bother with trans coding, creates copies of incoming streams and sends them to other participants “as is”. It turns out that each terminal receives several video streams in full quality at once, which it simply cannot display simultaneously. The terminal has to reduce the resolution of each incoming video from each of the participants on its side, or ask to reduce it before sending, which degrades the quality of the video for all other participants.
Plus, this scheme has one: the infrastructure is not demanding of resources and even an ordinary PC can withstand hundreds of such conferences at the same time. But there are much more minuses: the terminal (usually a simple PC) has to decode not one, but several streams at once, and the video conferencing server requires several times the outgoing width of the channel to accommodate all copies of the streams created by it.
Add to this the real conditions, and get a system that hardly “digests” the number of participants more than 3, and dramatically worsens the quality of the video for everyone, when a mobile subscriber who is not able to “digest” the original picture quality sent by others subscribers.
Scalable Video Coding (SVC) based video conferencing architecture
This architecture combines all the advantages of a mixing approach and, at the same time, is devoid of the disadvantages of systems based on multiplexing. It is cheap, instantly scaled and works on all platforms. This was made possible by the development of signal processing and data compression technologies.
The bottom line is that the terminal compresses its video stream in layers: each additional layer increases the video resolution, its quality and the number of frames per second. If the channel between the terminal and the VKS server is good, then the terminal sends the maximum possible number of layers. It is worth noting that the layer is not a separate video stream of lower quality, but a full-fledged difference between it and the previous layer. Thus, the SVC stream is only 15-20% different in channel width from a non-SVC stream, and is significantly less than the required amount of bandwidth of independent streams.
The VCS server, having received an SVC stream with layers, simply cuts off the extra ones without recoding, only by discarding data packets from it according to certain rules. Thereby, allowing on the fly to create individual sets of video streams (“layouts” of windows) for each of the participants of a group video conference, depending on its actual communication conditions.
Using modern protocols and codecs
For organizing video conferencing between various software and third-party equipment, standard data transfer protocols are used.
H.239 is a communication protocol supporting two media streams from different sources. Suitable for video conferencing in which the image is displayed on two different screens (for example, in the video conference room, when on one screen - the image of the speaker, on the second - the accompanying presentation).
H.323 is a protocol for transmitting data over networks with non-guaranteed bandwidth. It is used in personal and multipoint video conferencing.
SIP is a network protocol for establishing a connection between client applications from different manufacturers, replacing the H.323 standard. Used in video conferencing and IP telephony.
Compressing and playing audio and video during a conference call is done through the use of audio and video codecs.
- H.239 is a communication protocol supporting two media streams from different sources. Suitable for video conferencing in which the image is displayed on two different screens (for example, in the video conference room, when on one screen - the image of the speaker, on the second - the accompanying presentation).
- H.323 is a protocol for transmitting data over networks with non-guaranteed bandwidth. It is used in personal and multipoint video conferencing.
- SIP is a network protocol for establishing a connection between client applications from different manufacturers, replacing the H.323 standard. Used in video conferencing and IP telephony.
Compressing and playing audio and video during a conference call is done through the use of audio and video codecs.
- H.264 is a video compression standard that provides a high level of compression of the video stream while maintaining its original quality.
- H.264 Scalable Video Coding (SVC) is a codec with compensation for missing data that transmits video using multiple layers. Resistant to network errors such as packet loss.
- H.265 is a video compression standard that uses more efficient encoding algorithms than H.264. Among the features of this video codec, you can highlight the increased resistance to packet loss during media transmission and the minimum signal delay during video conferencing. This standard supports UltraHD formats: 4K and 8K.
- Opus - audio codec for audio compression, featuring high performance and scalability.
- G.722.1 Annex C is a compression standard for wideband audio.
- VP8 is a video codec with enhanced resistance to frame loss and high decoding speed of video streams.
- VP9 is an open video compression standard, the original purpose of which was to improve the performance of the VP8 and H.265 codecs. In the first case (compared to VP8), the main task of the developers was to reduce the bitrate by 50% while maintaining the original video quality, in the second (compared to H.265) - a significant improvement in the compression efficiency of video streams.
findings
When recommending a video conferencing system, we recommend that you carefully familiarize yourself with the principles of its operation and choose the one that will minimize the costs of its implementation, scaling and support. At the moment, such requirements are satisfied by all software video conferencing servers based on scalable video coding.
Source: Nexgen IT Solutions & Services
No comments:
Post a Comment