Most of the early breakthroughs in fixed/mobile convergence (FMC) have come from telcos bringing their DSL services closer to cellular, either with full SIP or an interim solution like Wi-Fi UMA. While some cable operators have taken an advanced interest in convergence as they seek to define their own quad plays, few have got beyond marketing alliances and some attempts at unified billing.
But the standards that will enable early movers like the US Pivot partnership to deliver a technologically integrated convergence are starting to fall into place, with the key body behind them, CableLabs, publishing specifications for voice hand-off between cable and cellular systems.
It has also started to receive responses to its RFI, issued about six weeks ago, for the all important release 2.0 of the PacketCable standard, which will take the operators towards full integration with other IP systems.
In June, CableLabs issued a detailed RFI for PacketCable 2.0, which is the successor to the VoIP-centric PacketCable 1.0 and to PacketCable Multimedia (PCMM), a platform that supplies quality of service for all forms of IP services. The emerging standard is based on IMS and borrows heavily from SIP, allowing services and apps to be bridged across wireline and wireless access networks.
CableLabs also recently launched a PacketCable 2.0 Applications Lab at its Louisville, Colorado facility. However, most cablecos do not expect to deploy PacketCable 2.0 services extensively until 2009 or later, and the new FMC specifications for voice hand-off may be more important to short term convergence moves ‘ in the same way that UMA has been in the telco sector.
Adoption of the SIP protocol and IMS ‘ the key standards underpinning service delivery over end-to-end IP networks – was always going to be smoother for companies that are already heavily IP-oriented, such as AT&T or VoIP specialists like Vonage; or for large cellcos, for whom the IMS standards were initially conceived.
But PacketCable 2.0 (Multimedia) will start to close the gap, allowing cable/wireless carriers like Canada’s Rogers to deliver on their promises of converged services, not just with unified bills and user interfaces, but with a common infrastructure at the back.
In addition ‘ and importantly for ventures like Pivot, which consists of the US’ four major cablecos plus Sprint Nextel ‘ there will be significant progress in 2008 towards roaming between different cable networks using IP.
The latest set of specifications from CableLabs defines how voice calls and other applications, such as call waiting, can hand off seamlessly between PacketCable and cellular networks using dual-mode handsets. The handsets and other such devices will offer a major opportunity for companies like Motorola, which already provide CPE to the cable industry as well as mobile handsets, and the giant is already looking to add Wi- MAX to the mix, no doubt with any eye on Sprint/Pivot.
This venture is a good example of how badly new cable standards are needed. Although it is very important for the US cable operators to offer mobile services in order to keep up with AT&T and Verizon Wireless in terms of the quad play, so far their agreement with Sprint is basically a fairly advanced co-marketing, cross-branding and shared billing pact, not unlike those of other MVNOs.
It brings none of the efficiencies of collaborating on a single umbrella infrastructure based on IP that can deliver to a wide variety of devices, over various RAN technologies ‘ something Sprint/Pivot would be able to do once there is effective integration between existing and future PacketCable systems, EV-DO (itself moving to IP) and WIMAX. At this point, the cost per Mbps should be at least halved, new applications deployed and provisioned more quickly and flexibly, and a host of new devices supported without the level of R&D investment that such changes currently require.
The voice hand-off specs are called ‘PacketCable Cellular Integration’ and define Voice Call Continuity (VCC) requirements between the two networks ‘ largely based on work already done by the 3GPP and 3GPP2 cellular bodies, but with some cable-specific enhancements. The main functions addressed are network domain selection and call continuity.
A VCC application server would determine how a call is routed and choose the available link with the best coverage. The platform’s ‘hybrid service execution’ component will allow cablecos to retain control of the customer and how applications are delivered over its own IP network or the cellular system. This adds to efficiency, by avoiding a situation where apps like call waiting are invoked in two different ways, but it also raises the question of control where, as in Pivot, the two networks are owned by separate and equally ambitious bodies. As with DSL, the most common mechanism in early deployments will be to link the cellular and cable networks via Wi-Fi, with a technology like UMA (Unlicensed Mobile Access), which could be extended to licensed spectrum IP systems like WiMAX or EV-DO in future.
Not covered by the specs is any roaming between cable networks, but CableLabs claims to be working on this advance. There is no date for official testing against the new specs, and they have to be seen in the broader context of the emergence of PacketCable 2.0 and the full all-IP multimedia architecture. However, some vendors, like CPE maker Arris, are already incorporating pre-certified implementations of the new specs into products such as home gateways.
For instance, Arris’ Touchstone WTM552 is a wireless residential gateway that incorporates PacketCable 1.0, SIP, a four-port router, Wi-Fi and two analog voice lines, and will later add 802.11e quality of service. Arris also recently announced its FMC product, the Mobility Application Server (MAS). All this is highly reminiscent of the telco world, with operators and standards bodies working on easily deployable convergence offerings, mainly focused around voice hand-off and dual-mode devices, that can provide a stepping stone on the way to IMS, but may also take the edge off carriers’ urgency to move to the full IP platform in the near future.
The large telcos are increasingly adding their own enhancements to IMS to make it more commercially deliverable, notably Verizon and AT&T, and in the mean time focusing on lower risk approaches that also offer full integration with legacy systems. The cable sector is taking a similarly cautious route. In a recent report by Cable Industry Insider, research firm Heavy Reading found that most cablecos would not move to full PacketCable 2.0 or IMS for at least 18 months, and sometimes as much as five years, and in the meantime would add SIP capabilities gradually.