BellSouth, AT&T Labs Execs Eye Optical Networks' FutureBellSouth, AT&T Labs Execs Eye Optical Networks' Future

Just days after their parent companies announced merger plans, senior executives from AT&T Labs and BellSouth Corp. provided a look into the future of optical backbones during keynotes at the Optical Fibers in Communications conference.

Loring Wirbel, Contributor

March 7, 2006

3 Min Read
information logo in a gray background | information

ANAHEIM, Calif. — Just days after their parent companies announced merger plans, senior executives from AT&T Labs and BellSouth Corp. provided a look into the future of optical backbones during keynotes at the Optical Fiber Communication Conference & Exhibition.

The irony of promising expanding traffic potential in a time of consolidation was not lost on OFC co-chairman Patrick Iannone as he opened the conference Tuesday (March 7) by quipping, “I’m from AT&T—but then again, who isn’t?”

G. Keith Cambron, formerly CEO of SBC Labs, was named senior vice president of AT&T Labs after SBC and AT&T merged in November. Cambron today is examining physical-layer implementations of Internet Protocol transport, particularly as first-generation broadband networks move to full IPTV capabilities. He told the OFC assembly that SBC’s Project Lightspeed is being broadened to encompass IP multimedia subsystem architectures that can handle wireless integration, home gaming and AT&T’s envisioned IPTV infrastructure, U-Sphere.

The combined SBC/AT&T global network now carries 5.1 petabytes of traffic per business day, most carried on multiprotocol label-switched IP networks with more than 1,500 ports. SBC did not bring in long-haul backbones, Cambron said, but it helped AT&T expand such last-mile access options as DSL. In the U.S. regional networks where SBC had operated, the AT&T Photonic Express backbone has connected states through OC-768 meshed dense wave-division multiplexed optical pipes.

AT&T will expand that network using photonic cross-connects that can unify existing Sonet and Ethernet traffic over IP. With the complex access network in place dominated by time-division multiplexed traffic, Cambron said, “the ROADM [reconfigurable optical add-drop mux] becomes an absolutely critical platform.” ROADMs can dispense with the need to install jumpers at intermediate transport points and eliminate “stranded bandwidth,” he said.

“In initial IPTV trials in one small region, we would have to do 30,000 individual fiber jobs without using ROADMs,” Cambron said. “And transition to Ethernet becomes easier with a ROADM infrastructure because we can react quickly.”

SBC began Project Lightspeed with ATM-based broadband passive optical networks connecting to VDSL in the neighborhood. This way, copper in the ground did not need to be replaced, but customers could gain bandwidth advantages beyond ADSL. As fiber buildout moves forward, AT&T will move to Ethernet-based gigabit PONs. As copper is removed from neighborhood networks, customers can transition from circuit-switched voice to voice-over-IP, Cambron said.

Henry Kafka, chief architect at BellSouth, used examples from the post-Katrina network rebuilding to make the case for preserving network resiliency. He noted that the storm had destroyed 32 offices in the Gulf region, affecting 1.2 million out of 20 million area customers.

BellSouth relied on “pre-WiMax” residential wireless service in the disaster area, offering customers DSL-like service over a four- to six-mile range, he said. Fixed wireless broadband access had previously been available only to small businesses but has since been expanded to residential customers, Kafka said. Where fiber networks have survived, they play a key role in backbone restoration. Where local broadband is required, RF is more reliable than free-space optics, Kafka said. Nevertheless, “wireless is in no position to replace fiber permanently, because of the large amounts of data we encounter.”

“If Katrina doesn’t drive next-generation network transport deployment, what will?” Kafka said. First-generation IPTV will be a largely multicast business using a broadcast and cable model, he said, but eventually customers will want multicast IPTV mixed with unicast and point-to-point Internet video, which drive up bandwidth requirements.

Kafka predicted that the access metric of the future won’t be bandwidth in gigabits per second, but capacity in gigabytes shared per month, reflecting the transition of IP video from passive transmission to interactive services.

Related link: Employ the best practices for managing optical and storage networks

Read more about:

20062006

About the Author

Never Miss a Beat: Get a snapshot of the issues affecting the IT industry straight to your inbox.

You May Also Like


More Insights