Matthew Goldman /
11.01.2005 12:00 PM
IPTV: Next-generation delivery of the television experience





Internet protocol television (IPTV) embodies next-generation technologies for delivering real-time and interactive television to the home or business, and it's available today.

This is not the low-resolution, sometimes jerky video that is commonly broadcast on the Web to computers. While IPTV does make use of Internet protocols and is carried over broadband networks, it differs from Internet video in that IPTV delivers broadcast-quality programming to a customer's TV over a network with quality of service guarantees. The general-use Internet has no such guarantees in place.

The technology behind IPTV is hidden from the viewer. From a customer's perspective, IPTV is similar to cable TV or direct-to-home (DTH) satellite service in the sense that he or she watches programming on a TV set, not on a PC, and selects channels using a remote control, not a computer mouse.

With the aid of an electronic program guide, the viewer selects a channel on the remote control and, from his or her perspective, that channel simply appears on the TV set. The viewer can enjoy linear real-time programming or engage in personalized interactive services, such as movies or other video-on-demand content, hard-disk type program recording, shopping and more.

While things may appear straightforward from a customer's point of view, the technology behind IPTV is anything but simple. There are four main elements to an IPTV system: a streaming headend, the core network, the last-mile access to the customer (fiber or a form of DSL) and the equipment located at the customer premises.

Streaming headend



Figure 1. A functional block diagram of a typical video headend architecture. Click here to see an enlarged diagram.



In many respects, the streaming headend resembles that for cable or DTH. Figure 1 on page 82 shows a functional block diagram of a typical video headend architecture.

Much of the programming, like national feeds of ESPN or HBO, is received via digital satellite and de-scrambled, but it is kept in the digital TV-compatible format in which it was transmitted. Other content sources may arrive via microwave, fiber or terrestrial antennas. A local studio can supplement the national feeds. These additional sources often require encoding of the programming to a DTV-compatible format. The ubiquitous encoding format used today for DTV is MPEG-2 video compression. The locally encoded and selected national feeds are aggregated and groomed to form an IP stream for delivery to the core network.



Figure 2. Some of the specific equipment—descramblers, IRDs, encoders, servers, IP streamers, switches—incorporated in the IPTV headend. Click here to see an enlarged diagram.



The headend also typically contains conditional-access and/or digital rights management systems (to control which subscribers are entitled to view selected content), VOD and other servers, streaming health monitoring equipment, and content insertion and automation systems. In addition, it can contain the portal or middleware software that handles presentation, services and authentication. Figure 2 on page 82 shows some of the specific equipment (descramblers, IRDs, encoders, servers, IP streamers and switches) incorporated in the IPTV headend.

The core network

To create the television experience viewers are accustomed to, the network must provide a signal that is free from glitches, lockups, dropouts and other problems. This is accomplished by ensuring that there is sufficient bandwidth and differentiated handling of services based on specified quality of service guarantees for the broadcast-quality streaming video.

Last-mile access

While the network may have wide bandwidth, the last mile to the customer premises may not. There are two last-mile technologies generally used in delivering IPTV: fiber to the home or business and digital subscriber line (DSL).

With fiber, bandwidth is not an issue. Fiber can easily carry MPEG-2 signals in both SDTV and HDTV formats. However, providing fiber to all of a telco's customers is a vast undertaking, which involves extensive redesign, construction and rewiring the drop from the street to the customer premises. Because of this, fiber is not as common as DSL.

DSL provides a broadband connection on existing twisted pair cabling, making life that much easier for the telcos by repurposing the existing drop wiring. However, DSL and its advanced forms are bandwidth limited. A typical residential asymmetric digital subscriber line (ADSL) service today may offer a bandwidth between 1.5Mb/s and 6Mb/s.

Bit rates vary with distance. If there is old wiring or wiring that is pigtailed or spliced in any way, the bandwidth will be reduced.

MPEG-2 is the ubiquitous encoding format today for both SD and HD. But in a network such as ADSL with constrained bandwidth, the bit rate for MPEG-2 is too high. HD in MPEG-2 requires roughly 12Mb/s to 18Mb/s just for the video part — not taking into account audio and data — and clearly won't fit into the bandwidth available with ADSL. SD in MPEG-2 requires approximately 2Mb/s to 4Mb/s. Therefore, the delivery of more than one channel to the customer premises (for viewing on a second or third TV) is not practical.

That's where the next generation of coding (compression) technology comes in. Both MPEG-4 AVC (ISO/IEC 14496-10/ITU-T Rec. H.264) and proposed SMPTE 421M VC-1 (industry standardization of Microsoft Windows Media Video 9) can code HD into 6Mb/s to 10Mb/s and SD into the 1Mb/s to 2Mb/s range, for equivalent picture quality to the higher bit rate MPEG-2. Now we're getting closer.

DSL technology is also leaping ahead. A new access network technology called ADSL2+ will allow bit rates (again varying by distance) in the 14Mb/s to 20Mb/s range. One channel of MPEG-2 HD could just about fit, but there would not be enough bandwidth to offer broadband service of support for more than one channel to be viewed on other TVs. However, a channel of MPEG-4 AVC or VC-1 HD — or multiple SD channels — will fit just fine.

IPTV is made viable by using these two technologies, which are coming of age at the same time. Advanced video coding approximately halves the bandwidth needed for HDTV and advanced xDSL more than doubles the bandwidth capability of existing DSL over twisted pair phone wiring. This benefits the service providers because they can now offer HD and multiple TV support to the customer and still have bandwidth left over for phone and data — the triple play package of bundled services.

Customer premises equipment and middleware

The customer premises equipment is usually a set-top box or a residential gateway. A key difference between IPTV and cable or DTH satellite is that the latter delivers all available channels to the customer at the same time. However, IPTV over a network with restricted bandwidth to the customer can't deliver, for example, 40 HD and 200 SD channels.

Using next-generation compression technology, with approximately 8Mb/s for each HD channel and approximately 1.5Mb/s for each SD channel, only a few channels will fit down a 14Mb/s ADSL2+ pipe. This means that the channel tuning function for IPTV does not reside in the set-top box, as it does for cable or DTH. Instead, channel switching takes place upstream in the network.

Switching can take place at the headend or somewhere else in the network, depending on network topology and the control architecture. In general, hierarchical control architecture is designed to optimize real-time response to the customer's push-of-a-button command on the remote control.

The control architecture that manages how users perceive and accesses available services provides the look and feel of the program viewing guide. It also verifies that users are authorized for the services they are attempting to access, which is commonly called the IPTV middleware or portal.

There are two aspects of the IPTV middleware that communicate with each other. One part resides in the headend or control aggregation points in the network, and the other part resides in the customer premises equipment.

In an IPTV system, when a user pushes a button on the remote, the remote sends a command to the set-top box. The set-top box then receives and interprets that command and translates it into higher level control signaling that gets sent upstream to the network channel-change controller and/or portal server. There, the command gets processed, with the result (like switching to a new channel) sent back down the network to the set-top box and onto the TV.

IPTV is here

IPTV is beyond the testing stage. Systems worldwide are already deployed. In the United States, many of the first tier telco companies will roll out IPTV services in full force in 2006 to better compete with cable and satellite. SBC plans aggressive rollouts to more than 40 metro areas, and Verizon has filed franchise applications in numerous jurisdictions. As content deals are announced, the service offerings are gaining momentum. Expect to see Verizon's FIOS and SBC's Lightspeed service offered in your neighborhood in 2006.




Matthew Goldman is vice president of technology for TANDBERG Television.



Comments
Post New Comment
If you are already a member, or would like to receive email alerts as new comments are
made, please login or register.

Enter the code shown above:

(Note: If you cannot read the numbers in the above
image, reload the page to generate a new one.)

No Comments Found




Thursday 10:05 AM
NAB Requests Expedited Review of Spectrum Auction Lawsuit
“Broadcasters assigned to new channels following the auction could be forced to accept reductions in their coverage area and population served, with no practical remedy.” ~NAB


 
Featured Articles
Discover TV Technology