Subscribe / Unsubscribe Enewsletters | Login | Register

Pencil Banner

To get the most out of WebRTC, integrate it into existing collaboration tools

Ashan Willy, SVP, Worldwide Product Management & Systems Engineers, Polycom. | June 24, 2014
The promise of WebRTC is the mass adoption of voice, video and file collaboration, but how does it fit within existing communication systems?

The International Telecommunication Union predicts Web Real Time Communications (WebRTC) will be installed in more than four billion devices by the end of 2016. Indeed, the promise of WebRTC is the mass adoption of voice, video and file collaboration, but the question is, how does it fit within existing communication systems?

Drafted by W3C with protocol work done by IETF, WebRTC simplifies the complex world of real-time communication. Although WebRTC isn't confined solely to web applications, embedding real time communications directly into Web browsers has been the focus for most of the industry. After all, the promise of free high quality video and audio in one of the most deployed application in the world (Web browser) is exciting. As browsers work to integrate and refine the technology (Firefox and Chrome support it today), WebRTC holds the promise of instant connectivity through familiar interfaces with little or no additional software.

WebRTC defines browser APIs together with a collection of communication processes and protocols. From a development perspective, core functions are encapsulated into three main JavaScript APIs: getUserMedia, RTCPeerConnection and RTCDataChannel. These APIs are incorporated into browsers that support WebRTC, hence a web developer who has JavaScript programming experience can bring an interactive video collaboration experience to the web. The following diagram illustrates the architecture:  

Roadblocks to scalability
Like any new technology, WebRTC isn't without issues. Traditionally, video and other collaboration systems have used conferencing bridges to link a certain number of participants to a call. WebRTC technology works by changing the structure of the connections between parties.

WebRTC allows a mesh-based technology to enable users to send and receive streams to and from each other. This is not new in the video world, as technologies exist that accomplish this today. Each stream operates independently, which reduces the strains of conferencing applications (as bandwidth doesn't aggregate to a single choke point) unless, of course, bandwidth inefficiencies come into play.

With this mesh technology, WebRTC can, in theory, accommodate an infinite number of participants on a call. But in practice, the more parties that join a call, the more bandwidth that call will take. Bandwidth inefficiencies can mount quickly, as each device connected to the call receives and transmits multiple transmissions. If bandwidth falters, quality suffers and the entire call can fail.

On these more complex calls, signaling factors in as well. In the past, Session Initiation Protocol (SIP) has provided a way to register users and identify them uniquely, as well as to manage call notifications and modifications. WebRTC in its infancy does not include a concrete means of signaling, leaving some basic call functionality up in the air. Without protocols for connecting, disconnecting and identification, disorder can ensue.

 

1  2  Next Page 

Sign up for CIO Asia eNewsletters.