Articles
The Society of Cable Telecommunications Engineers' Conference on Emerging Technologies saw a flood on many levels this year—not only from the weather, but also from attendee numbers and the array of technical information presented during the three-day event.
 |
|
SCTE’s ET ’05 drew 935 attendees
to Huntington Beach, Calif. |
The conference in Huntington Beach, Calif. began even as the region was hit with historic rainfall and flooding. But as the deluge subsided outdoors, a torrent of a different kind began indoors—one of technical information in the form of white papers and panel sessions filled with observations, discussions and even disagreements on the latest in cable technology.
The conference, which attracted 935 attendees (up 20 percent from the '04 gathering in Dallas), covered topics ranging from analysis of competing technology to digital rights management, home networking, wireless systems, bandwidth management and converged network architectures.
The journey to an all-digital world where entertainment and communications will easily flow across many devices won't be an easy one, but computer pioneer Paul Allen, who keynoted the conference, said it is one in which cable is poised to take the wheel and drive.
Allen, co-founder of Microsoft Corp. and owner of Charter Communications Inc. parent company Vulcan Ventures, took the crowd through a wide-ranging roadmap of future services, telling cable's engineers that "we're in the driver's seat; we have the superior platform," to create and field the multimedia and communications services consumers will want. But he also noted the challenge lay in pulling together wired and wireless technologies for data, television, voice and mobility—the latter ranging from Wi-Fi to cellular systems—via a common digital network base.
"We have to have great operations in the field to make all of this work together," Allen said. "We have to figure out ways to make that happen, with seamless integration."
At the same time, cable operators also face challenges from the galloping rate of innovation. "Every year we see the technology side of the products become better," Allen noted, but he also warned that "one of the things we have to internalize is these things will be happening faster, and at a rate we haven't seen in the past."
That phenomenon extends to good-old telephony, where VoIP is fast developing to include video. The number of things one can do with the technology "are pretty amazing," Allen said. And cable data services are now able to lash together multiple data channels to offer hundreds of megabits per second to customers.
Key to developing all of these services is the migration to an all-digital platform. Pointing out that Charter's Long Beach all-digital deployment was a step in this direction, he advised that once cable operators have moved a majority of their customers to all-digital transmission systems, "you can find ways to subsidize moving the rest to digital."
If cable operators can overcome the inherent hurdles, they should be well positioned to offer the best vehicle for next-generation services in the future, he concluded.
Despite the challenges "this is going to happen in the next five years, and that is very exciting," Allen said.
Banking on the advice that it is best to know one's enemy, cable engineers in another session picked up technology reconnaissance reports on competing technologies ranging from telco-centered fiber-to-the-home systems to emerging wireless schemes such as metro WiMAX and upgrades to satellite platforms.
Chief among those competitors are telcos fielding DSL-based technology. David Reed, chief strategy officer at CableLabs, alluded to the rain deluge in the Los Angeles area during the week, saying he had no intention of "raining on cable's parade" with his views on DSL technology.
SBC has discussed ADSL2+, while Verizon has stated it will move directly to fiber-to-the-premise (FTTP) to deliver video service, he noted. ADSL2+ does improve DSL's throughput, but the technology still requires 8,000-foot loop lengths to deliver the necessary bandwidth.
At the same time, new broadband wireless technologies such as WiMAX and metro-area extensions of Wi-Fi are also coming into play. Sandy Teger, co-founder of System Dynamics Inc., noted that for cable operators the two technologies weren't a cause for panic, but it was important for them to understand the implications.
WiMAX is developing in metropolitan area networks as a more or less fixed service, with products expected this year. But with development of the new 802.16e standard to add call handoffs between base stations, WiMAX will become a mobile access technology, Teger said.
"That's the piece that all of us have to be aware of and figure out what to do, because this is what the customer wants," she noted.
Steve Osman of SES Americom updated the group on the latest in satellite technologies, including potential upgrades that could give DBS providers HD capacity that rivals cable's. That includes the evolution of multi-beam satellites combining Ku- with newer Ka-band spot-beams and advanced frequency re-use systems that could boost capacity from two to 20 times.
On top of that, satellite providers can field "tweener" satellite positions 4.5 degrees from existing slots, as well as Ku- and Ka-band "topper" birds riding above the traditional orbital arcs. That could further double capacity, Osman noted.
With those tools, and new compression codecs such as H.264/MPEG-4 that can cut the HD bandwidth load in half, satellite competitors could raise their HD channel capacity to virtually match that of a cable operator's 860 MHz plant, Osman said.
If cable operators continue to support analog channels or don't similarly convert to advanced compression codecs, "they will be at a distinct disadvantage," he surmised.
A series of experts looked at the issues that play into turning a simple best-effort data service into a more sophisticated, complex means to deliver premium services while controlling other usage and protecting against predatory application worms and viruses.
Susie Kim Riley, founder and chief technology officer for Camiant Inc., focused on dynamic bandwidth management used in cable networks.
"Things are starting to converge at the network, and applications are also converging as well," Riley said.
Enter PacketCable Multimedia, which will allow operators to apply QoS to target applications. A key there is in the policy server, which allows the operator to control how network assets are tapped by specific application, Riley noted.
But the cable network—at least for now—is essentially a dumb pipe, according to Milind Gadekar, senior manager of service control platforms at Cisco Systems Inc. After first apologizing to the cable engineers for that characterization, Gadekar told them that to change dumb pipes into smart ones requires four key elements: service control, subscriber management, real-time traffic control, and staying up with new applications.
"In today's broadband network it becomes more complex because applications change and morph, and the network has to be able to keep up," Gadekar said.
With all of these changes looming for networks, the cable modem termination system (CMTS) also will have to evolve. Tom Cloonan, chief technology officer for ARRIS' broadband division, told the crowd there are plenty of new features on the horizon to help operators manage bandwidth.
Cloonan said it is getting harder to figure out the bandwidth needs of subscribers because the applications they choose to tap are becoming less predictable. But DOCSIS 2.0 provided an array of tools to change bandwidth usage, which can save as much as 30 Mbps in a plant, he added.
Cisco's new wideband protocol scheme, meanwhile, also is aimed at making the next-generation cable network more efficient, according to John Chapman, a Cisco distinguished engineer. By eliminating the DOCSIS service subchannels and creating a single wideband channel, cablers can gain significant bandwidth.
"There's nothing simpler than a big, fat pipe," Chapman said. "That's what we are trying to do with DOCSIS—is create one big, fat pipe."
A related strategy calls for bonding data channels so that they share traffic load but retain their channel boundaries. Thomas Quigley, senior director at Broadcom Corp., said there are several strategies available for channel bonding, which essentially bolt together adjacent data channels to boost overall traffic efficiency. Because it retains the channels' integrity, it can be put in place without disrupting legacy network systems.
But as with everything, there is a downside. For channel bonding and wideband, the penalty comes in recombining data at the endpoint, and, for wideband specifically, that penalty extends to modifying cable modems to recognize the bigger data channel. Because the bonded channels or the single wideband channel are delivering a much larger load of data, it requires more sophisticated recombining to sort out the flood of data and make adjustments for variation in packet arrival time.
"Certainly, as we define these specifications at CableLabs and in the industry, that is something that we need to be aware of," Quigley noted.
Getting a fix on switched broadcast
The idea behind using switched broadcast technology is simple: by "switching"—instead of broadcasting—all or some of the channels in a given lineup, it follows that an operator stands to realize significant bandwidth savings. However, most of that discussion has been theoretical. Until now, that is.
Ran Oz, the CTO of switched broadcast evangelist BigBand Networks, and Nishith Sinha, iTV systems engineer at Cox Communications, shared the findings of two "proof of concept" trials—one with Cox and another with an undisclosed MSO.
During those trials, BigBand and its partners uncovered several key statistics, including how much bandwidth an operator might expect to save in different scenarios, as well as some valuable insight into the viewing habits of digital cable subscribers.
"Trial A," conducted with Cox, was done on a spectrum-limited 550 MHz system in Tyler, Texas. Trial A, which eventually reached four nodes, switched 60 digital channels already available as a broadcast tier. "Trial B" took a broader view by switching the entire digital lineup, sans high-definition and music channels. Trial B eventually reached four nodes and topped out at 171 switched channels.
In Trial A, it was discovered that channel usage peaked at 50 out of a possible 60 channels—not a huge bandwidth savings, but the results were likely "artificially higher" due to set-top "tune in" and "tune out" messages.
Trial B, however, found that substantial bandwidth savings could be made when a greater number of less-popular channels are included. There, the maximum number of channels ever used never exceeded 67 of 171 available—a bandwidth savings of 60 percent.
Home networks may connect cable operators to new and lucrative services, but they also present potentially vexing problems with device interoperability and digital rights management.
A panel of experts tackled these technical issues, and while there were few clear answers, there was seemingly no shortage of possible avenues cable operators could take.
Doug Jones, chief architect at YAS Broadband Ventures, walked cable engineers through issues surrounding conditional access in the future home multimedia network.
For cable, conditional access is the basis for digital rights management, Jones said. In the past, that has governed content streaming from the headend to the home, but that has changed with DVR technology.
"Now that we are dealing with stored content, we are dealing with playback and recording issues," Jones said. "It adds specific usage rules of the content."
In the future, it may also be possible to move content between cable conditional access keys. That would allow customers who move from one cable system to another to transfer stored content between the systems, even if they didn't use the same conditional access schemes.
Indeed, for RealNetworks, the vision of the future home could allow content to be transferred from a home DVR player to a server in a car via Wi-Fi, allowing passengers to continue watching the content during a trip, according to Jeff Ayars, Real's general manager of embedded players and client technologies.
Interoperability of DRM systems to secure that content will be key, and Ayars noted there are at least three groups working toward that goal. The IEEE is developing MPEG-21, while the European Union is working on an overarching DRM overlay dubbed OPERA. But Ayars focused on the Open Mobile Alliance's work on a DRM 2.0 specification aimed at content streaming to mobile phones, now ready for implementation.
The OMA DRM standard—which Real had a hand in developing—was intended primarily for cellular phones, but "the specification itself is significantly robust and it was designed in order to support high-quality, high-value content," Ayars said.
Turning homes into miniature VOD servers also is part of the future, according to Gary Hughes, director of media engineering at Broadbus Technologies.
By 2008, more than 50 percent of homes will have home networks, and home media services will be a central feature, Hughes said. With the evolution of devices able to access, store, index and distribute content, "it's all sounding rather like a home mini-headend," he noted.
New services such as VOD magazines and "hyperchannels" offering specialized content are also on the horizon, but that will bring along thorny content protection issues. In particular, Hughes noted metadata could pose problems, with the need to extend program information beyond just the bullet description in the program guide.
From a consumer electronics perspective, interoperability and consumer behavior patterns are key, according to Paul Liao, chief technology officer for Panasonic Corporation of North America. While digital media has caught on, people are still moving content around via physical data files such as CDs, Liao noted.
In response, Panasonic is looking seriously at powerline home networking. Not only does it require no new wiring, but there are also plenty of connection points in the power outlets.
"While there are lower-speed products out on the market, what we are really looking for is higher speed, at upwards of 100 megabits," Liao said.
In the end, Liao said that consumer multimedia devices need to protect content but be easy for consumers to use and hopefully offer a better experience then they do now. On the downside, the danger is in trying to alter consumer habits.
"If you try to change consumer behavior—good luck," Liao said.
|
