Rethinking Transport for the Tele-Entertainment Age

The two previous instances of this blog elaborated the causes behind the failure of ATM to become the convergent technology. Before we look at the candidate technologies lined-up in the starting blocks, it may be worthwhile to first examine the services to be supported by these same technologies.

The first point that should be made is that high-QoS, connection-oriented packet traffic used for tele-entertainment is rapidly supplanting the basic data packet traffic as the bandwidth driver in the global networks. As a matter of fact, over-the-top video traffic has passed peer-to-peer traffic as the single user of bandwidth.

At the turn of the century, network service providers poured money into data networks and settled themselves into a convergence strategy optimized for the support of large quantities of connectionless, non-real-time data services with a small amount of connection-oriented voice traffic poorly handled via bandwidth over-provisioning. Just a few years later, they suddenly find themselves struggling to handle new interactivity models that demand quick delivery of bandwidth-hungry video traffic.

The same advances in A/V compression that have enabled the transport of A/V services over IP are further disrupting the stressed networks because current compression protocols work best when encoding video as variable bit rate (VBR). This, in turn, results in unpredictably bursty traffic patterns since the bit rate is now tied to a variable video content.

The impact of this bursty video traffic is three-fold:

1. It causes bandwidth waste when fixed-bandwidth channelization techniques are applied to isolate particular video streams from other traffic.
2. It invalidates the packet-mode QoS methods traditionally used for the transport of constant bit rate (CBR) traffic.
3. It increases packet delay variation (PDV) in the network due to the queues filling up with the bursty traffic, further impacting the quality of packet-based voice.

Given the bandwidth requirements of video and the connection-oriented nature of these tele-entertainment services, doubts are being cast on the effectiveness of the architecture of these freshly-deployed data networks in handling these A/V services in an adequate manner.

The telecommunications community as a whole appears to be only reacting to these facts and still seems to be trying to assess the magnitude and implications of these changes. For now, they are dealing with it by over-compressing the video content while trying to assess the tolerance level of the customer to pseudo-HD IPTV.

However, some yet newer services such as HD video conferencing, networked data center support, and remote video gaming, just to mention a few, are just coming upon us with even higher QoS requirements.

Even legacy applications, such as conversational communications, that had thrived virtually unchanged throughout the last century, may become even more demanding with the possible addition of video calling, for example.

Finally, cellular networks are spreading into the residence, in the form of femtocells, bringing along with them timing and synchronization requirements that were never foreseen during the data network explosion.

The common thread here is a combination of video-driven bandwidth and interactivity. Pre-packaged content sources such as broadcast TV are being largely bypassed by a mobile generation intent on getting instantaneous access to dedicated content. Web page reading and e-mail, while still needed, will have a diminished role in this new world order. So will poor video quality.

The motion picture and consumer industries are marching inexpugnably toward higher resolution formats and faster frame rates. While the so much heralded IPTV revolution is still figuring out how to stream a single high-quality HD video stream, Samsung is touting its 63-inch 4k x 2k plasma, JVC is showing off 8k x 4k D-ILA projector, and the Hollywood studios are trying to get their hands around an industry-standard 3D format.

How we will connect these new technologies to the correct tele-entertainment sources is still an unanswered question. Everybody wants to be a content provider, but no one has yet figured out how to deliver electronically that same content in a reliable and timely matter. All we keep on hearing is the lame argument that there is no money to be made in transporting multimedia entertainment content. If this is true, maybe some of these MSO CEOs would like to trade houses with me.

Alternatively, could it be possible that contemporary telecom transport got itself trapped into a corner while perpetuating antiquated technologies that cannot address current needs and even less the upcoming services described above?

We appear to be spending an inordinate amount of time, effort and money trying to shoe-horn legacy technologies that never envisioned these requirements and may never be able to support them effectively.

One would like to believe that we still have more innovation left in us beyond rigging unsuitable protocols to do a job that in reality begs for new thinking from the telecommunications industry.

Serge Fourcand is a principal engineer with Huawei Technologies USA. He is an accomplished product manager and system architect with extensive multimedia experience in both consumer and telecom industry segments, including end-consumer multimedia products and associated controls, advanced multimedia distribution, speech processing, IP-based video and audio transport, digital signal processing, Ethernet Layer 2 transport and switching, and end-to-end telecommunications technology. Prior to Huawei, Fourcand worked as a product manager for AMX Corp., senior hardware architect at Metro-Optix, a system architect for Broadband Gateways and Alcatel USA’s Switch Products Division, and has been a member of the technical staff of Siemens AG in Munich, Germany.