Wireless Watch: Small cell sector fights to bring its multivendor dream to carrier networks Jul 5, 2012 – Rethink Research
The residential femtocell may still constitute the vast bulk of small cell sales (86% of a total of 6.4m this year, according to Informa estimates) but if last week’s Small Cells World Summit was anything to go by, it is already considered rather old hat. The product launches, carrier roadmaps and hot debates all centered on public access small cells, whether outdoors on lamp posts or inside venues – even the Femto Forum has rechristened itself Small Cell Forum. Yet the humble femto has provided many of the concepts on which the new wave of complex, carrier class metrocell networks will rely, and by its very existence has shifted much of the thinking of the RAN industry. The challenge now is to ensure those shifts continue to influence the development of the whole carrier infrastructure, bringing new norms to the public network where the established powerhouses are still defending their view of the world.
In particular, the femtocell ushered in three disruptive notions - that a base station could be built around a low cost, integrated system-on-chip that looked more like the guts of a handset than a piece of infrastructure; that a product operating in the carrier network could be self-installed and self-managing without disaster occurring; and that operators really could mix and match access points and gateways from different suppliers, avoiding the dreaded lock-in. As such, the femtocell brought cellcos some of the advantages associated with Wi-Fi infrastructure, but without the downsides of unlicensed spectrum. No wonder carriers were more proactive in driving initial small cell standards than they have normally been in other areas.
The question now is, having shown the way to a new thinking about wireless networks, can those femto benefits really be transferred to the more complicated and mission critical world of the public RAN? The incumbent suppliers can live with a home and even enterprise world of low cost, tiny base stations that lend themselves to the economics of the consumer electronics vendors and Taiwanese supply chain, provided they maintain their power base in the main carrier RAN and core. Some actively embraced the private femtocell wave, providing gateways and even the access points themselves, like Alcatel-Lucent. Others, like Ericsson, stayed aloof, and were always swift to draw a very thick line between the femtocell - a handy but somewhat trivial consumer device, went the argument – and the ‘real’ network. That line is now breaking down, and letting the ideas of the femtocell into the main carrier RAN is deeply disruptive. We can expect a far stronger pushback against commoditized and multivendor base stations than we saw in the first femto wave.
Predictably, the willingness of large OEMs to embrace small cells is in inverse proportion to their incumbent strength in macro networks. So NEC, which was pushed out of 3G infrastructure outside Japan, and Samsung, whose RAN base lies in WiMAX, are keen to mount a rearguard action for LTE based on small cells. Some, like ALU, are treading a middle ground, recognizing that they could use multimode small cells to infiltrate their rivals’ larger 3G deployments, but eager to convince operators that sourcing all those metrocells from a single vendor, perhaps with managed services to match, would avoid all kinds of headaches.
But the multivendor aspect is perhaps the closest of the femtocell’s three disruptive qualities to the hearts of many carriers. The biggest psychological, if not technical, barrier broken by the Femto Forum was the agreement of the heavily operator driven Iuh standard interface, which links the cells themselves to gateways in the operator’s core network. This replaced proprietary interfaces and allowed femtos from many vendors to be included in one system, driving competition. This concept, commonplace in the PC/IP world, was revolutionary in a mobile industry which had always been based around proprietary connections and closed, end-to-end networks.
However, bringing Iuh – or the alternative standard link, SIP – into the public metrozone in place of traditional Iub is only one step to preserving the mix-and-match dream. ALU, for instance, supports Iuh in the public network but still insists end-to-end systems will be the best option in large scale roll-outs because other standards are inadequate, especially when it comes to interoperability between the small cells and the macro layer.
This is where the metrozone introduces a host of new issues, many of which will only get worse with LTE-Advanced, which defines new features which will need to interwork. To ensure full quality of service, rather than merely coverage and capacity, the macro and small cells need to coexist happily – or create a “new material” as Nick Johnson, CTO of small cell specialist ip.access, put it. Will Franks, his counterpart at rival Ubiquisys, agreed on this point at least, saying: “The challenge is no longer data rate or call capacity but KPIs in variable traffic such as cell overload, or data versus signalling.” That means the new capabilities which LTE and LTE-A introduce to enhance self-organization must be harnessed, but they require full harmony between all kinds of different cells present in a zone.
As Cisco’s Mark Grayson, a distinguished consulting engineer for mobile architectures, put it: “When we invested in ip.access [in 2008] we thought we could do residential roll-outs with no macrocell integration but we’re pushing the envelope of that assumption now. Once you move from tens to hundreds of access points, even in residential you need macrocell integration.”
That takes companies like Cisco out of their comfort zone. Cisco’s strengths lie in large private and controlled enterprise networks, based on IP and starting to incorporate femtocells alongside Wi-Fi and wireline. It is even trying to usurp some of the roles of the service provider in the home, by extending applications and management tools from the cloud to its Linksys consumer access points. But when it comes to the mobile network, it has stuck to its strengths in the core, and avoided clashing with the major OEMs in the RAN (except for a brief and ill-considered flirtation with WiMAX). But now, like the smaller femtocell specialists, it will have to work alongside the incumbents of the public access network, and they will not be opening the doors wide to welcome in the newcomers.
On paper, the key weapon of openness, once the small cell has to cohabit with the macro network, should be the 3GPP’s X2 standard. This interface supports communication between base stations to support various functions such as handover, load balancing, CoMP (coordinated multipoint signals to reduce interference), cell activation and organization. X2 standardizes these control plane procedures to make different vendors’ eNodeBs interoperable. However, not only does X2 in its vanilla form add to bandwidth and latency, but it is inadequate for complex networks of many cells.
It is too vague in its basic form. For instance, one technique to avoid interference is for macro signals to use ABSF (almost blank subframes) to transmit to small cells. This means that one layer does not transmit L1 control signalling within given subframes. However, the standard does not define which blank subframes to use, so two small cells can clash while trying to transmit in the same ones.
Such weaknesses mean that vendors are already adding proprietary extensions or interpretations. Grayson commented: “Standards can define the X2 protocol but vendors can deploy it with different algorithms so it isn’t interoperable. CoMP and other features are moving back towards proprietary.” This is driving the Small Cell Forum to set up a taskgroup to examine the additional capabilities that are needed, but in a unified way – as with Iuh, it would then submit the proposals to the 3GPP. This is likely to meet with sterner resistance than when femtocells were confined to private access and could be viewed largely in the same category as Wi-Fi access points. Grayson warned: “There is a pushback from the tier one OEMs not to go beyond X2.”
Johnson outlines the challenge more starkly. Interlayer interoperability is essential and standard X2 signalling is not enough, so operators can either compromise and live with a single vendor, or if they want to cling to their multivendor dream they have to think about these problems, especially in Release 10. “Operators can accept a de facto RAN monopoly in their networks, or support these interoperability standards,” he said.
Publicly, large carriers like China Unicom and Vodafone are heavily supportive of new standards efforts. Yet other operators, more privately, express fears that, in the end, the single-vendor solutions to which they are accustomed will just prove too alluringly low risk, when there are so many network decisions to be made. The Forum, nonetheless, will certainly have heavyweight support from operators and non-incumbent vendors like Cisco. If it can make its recommendations as swiftly as it did with Iuh and other early femto technologies, it will stand a good chance of ensuring that carriers can choose the multivendor route in metrozones too. If the new standards do not take off, there will be little impact on cellcos’ enthusiasm for small cells. And some of the other defining qualities of the femtocell will still find their way into the carrier network as a whole, notably the SoC-based base stations and many elements of self-management. However, a significant opportunity to push the mobile world further towards the openness that should accompany all-IP will have been lost.
Wireless Watch: Small cell sector fights to bring its multivendor dream to carrier networks
