Cable Moves Closer to Capitalizing On Efficiencies of PacketCable 2.0

Chris Busch, VP, broadband technology, Incognito

Chris Busch, VP, broadband technology, Incognito

September 30, 2009 – Not much has been heard lately about MSOs’ plans for deploying the next-generation communications capabilities of the PacketCable 2.0 protocol, but, below the radar, cable’s accelerating push into enterprise and mobile services is generating much work around preparations for applications tied to the new specifications.

One sign of this activity is Incognito Software’s release of support for PacketCable 2.0 in its device provisioning suite, known as Broadband Command Center. By supporting this standard, Broadband Command Center allows service providers to reliably provision PacketCable eDVAs (embedded digital voice adapters) on a network-wide basis and to maximize their core investment in PacketCable technology, says Chris Busch, vice president for broadband technology at Incognito.

“By releasing the PacketCable 2.0 provisioning capabilities at this point we’re letting operators know that the support is there for implementing the specifications in new devices that are shipping now, such as DOCSIS 3.0 modems with embedded voice adapters,” Busch says. “In the approach CableLabs has designed for the upgrade path, you can ship these devices now as PacketCable 1.x eMTAs (embedded multimedia terminal adapters) and then later signal the device to receive the new firmware image that will turn it into a 2.0 eVDA.” The PacketCable 2.0 capabilities in the Incognito release support the provisioning of such firmware upgrades, he notes.

The PacketCable 2.0 standard allows operators to move beyond the cable-specific voice-over-IP platform established by the original PacketCable regime to the pan-industry VoIP environment that has emerged with adaptation to SIP (Session Initiation Protocol), where SIP-based servers and clients can set up communications with each other irrespective of what types of networks they reside on. PacketCable 2.0 also embraces IMS (Internet Protocol Multimedia Subsystem) technology, which is designed to extend the cross-platform convergence capabilities to all IP-based data, including video and text as well as voice.

In April CableLabs certified the first PacketCable 2.0-compliant eDVAs. Such devices, in this case supplied by Ambit and Thomson, are designed to work as cable modems and as VoIP adapters that are compatible with SIP. They’re also compliant with IMS aspects of the standard, but, as Busch notes, much work remains to be done before the IMS convergence capabilities can be practically implemented.

For now, the focus is on SIP, especially in the commercial services arena, where SIP-equipped clients such as IP PBXs and SIP-based hosted PBX services have become commonplace. Cable operators seeking to serve these businesses with commercial voice services have had to use SIP servers or PacketCable-based softswitches that are SIP enabled but not PacketCable 2.0 compliant, which complicates integration of VoIP services onto a single operating platform.

“Operators have deployed non-IMS, non-PacketCable 2.0-based SIP softswitches in their networks just for business services in order to get market share,” Busch notes. “But, going forward, they don’t want to continue to be creating islands of voice network services which then have to be glued together with gateway session controllers or whatever. They want to offer native SIP trunks in a fashion that’s consistent with PacketCable 2.0.”

Even where PacketCable 1.x-compliant softswitches with SIP connectivity built in are deployed the architectures are highly limited for commercial service applications, Busch says. “With legacy PacketCable 1.x architecture we didn’t have a native SIP trunk,” he explains. “We had a PacketCable SIP trunk where Cablelabs defined the ability for two PacketCable 1.x call servers to talk to one another over a SIP-based trunk. The net result of that is if I want to run a SIP trunk off a PacketCable 1.x softswitch, I can only talk with another 1.x softswitch. Having a voice core offering native SIP trunk support that’s inherent to its architecture is imperative to addressing IP PBX service requirements.”

Beyond specific service applications there are over-arching strategic advantages having to do with operations efficiencies and major cost savings that are motivating cable operators to gear up for PacketCable 2.0. As Busch notes, once an operator implements PacketCable 2.0, peering across networks on a switch-to-switch basis only requires that all switches be compliant with the IMS/SIP-based 3GPP (Third Generation Partnership) architecture as distinct from requiring they be PacketCable compliant. “Now we have a PacketCable voice architecture which follows a larger industry set of standards for signaling and inter-application functionality,” he says.

This opens a way to save money, for example, in connecting calls between remotely located cable systems, because with PacketCable 2.0 operators can natively switch calls without having to go through external media gateways to a telco rate center and then through another gateway to the cable system attached to a second rate center. Similarly, use of PacketCable 2.0 allows cable operators to set up bilateral relationships with, say, a local exchange carrier that has put a native IMS core in place. “Neither has to put media gateways between each other to transfer calls from PacketCable to SIP,” Busch says. “Now it’s just SIP-to-SIP over a standardized SIP trunk.”

Another early driver behind PacketCable 2.0 implementation is cable mobile services, where operators want to be able to seamlessly hand off voice communications between VoIP-based mobile and home broadband VoIP connections. Here again the common link is the SIP connection.

Eventually operators will want to implement the full multimedia convergence capabilities implicit with IMS. “In the future we’re talking about Incognito handling provisioning to a terminal that happens to support voice as well as presence,” Busch says, in reference to the ability of any device to access any service from any location. “None of that will happen without the rollout of NGN {next-generation network) architecture, which is what PacketCable 2.0 represents.”

However, before the industry can get to that level of convergence much work needs to be done.
“Follow-me applications from the laptop to mobile to other devices are wonderful,” Busch says, “but where that ties into device provisioning is as yet undefined. And how that generates revenue is a bit of a mystery.” To get there requires that service categories be seen by the network as applications rather than as video, voice or data products, he notes.

“How do I enable my mobile phone to see the same content another device is viewing?” Busch asks. “What if each one of my video devices is going to support call display or even act as a voice terminal?

“In the network I have to register them as voice terminals and lines,” he continues. “I have to know how many need provisioning. I’m talking about hybrid IP voice and video with different paths and signaling, and those devices have to be provisioned for. There is no CableLabs standard for provisioning that now.”

But to address provisioning in a way that suits the needs of consumer-purchased convergent devices requires cooperation beyond the discussion that goes on inside CableLabs. In fact, given the progress already underway in this domain at the Broadband Forum (formerly the DSL Forum), it might be necessary for cable folks to talk to the telco people about such matters.

Sources say some preliminary reaching out in this direction is already underway. Whether it leads anywhere remains to be seen, but it’s hard to see how the convergence dream gets realized without such pan-industry cooperation. “This is the moment when we need to see some sharing of standards on a best-of-breed basis,” Busch says.