Small Cells World Summit 2014 Event Report

SCWSLarger volumes, increased delay, congestion, longer queues forming... perhaps Small Cell World Summit was trying to give a taste of possible future network problems when delegates first turned up early for this year's annual industry jamboree. In fact it was down to a temporary power outage at an unfortunate time, temporarily preventing the front desk from issuing badges. Quickly solved, the rest of the event ran very smoothly.

ThinkSmallCell was heavily involved – chairing two morning sessions, presenting and on the judging panel for the Small Cell Forum awards, together with many meetings – which delayed our reporting of the event. But it's also become a more complex and comprehensive marketplace, making it worthwhile to take a few days to cogitate and prioritise the key points rather than just regurgitate all the press releases.

An impressive operator line-up from the outset

The opening panel session assembled an impressive range of European CTOs who all professed support for Small Cells. Telefonica O2 see Small Cells as the right solution for densification of Urban Areas. UK operator 3 revealed they already had 100,000 3G small cells in operation (I think mostly residential 3G). EE said Small Cells were really useful tactically during their major RAN consolidation project, when they decommissioned many duplicated macrocell sites. Portugal Telecom contrasted deployment in their home country, where fibre is available 98% with that of Brazil, which has relatively little and their own brand Oi has different set of challenges. Those operators with both fixed and mobile networks were thought to have a long term advantage.

Clarifying the different Small Cell architectures

Delegates might be forgiven for thinking there is a much greater problem to solve in figuring out how best to deal with upcoming demands on their network. Terminology has often been an issue when discussing Small Cells, and some of the latest solution architectures only seem to add to the confusion. Mike Schabel, VP Small Cells at Alcatel-Lucent, did well to clarify the various options from Cloud RAN to traditional Small Cell. It's all about where you put the radio processing and baseband processing. DAS does both centrally, Small Cells do both within each remote node, DRS (Distributed Radio Systems) leave the RF at the remote node, but consolidate the baseband processing centrally. Each configuration has it's own sweet spot. He cautions the industry from confusing the market too much as it continues to innovate further.

The chart below illustrates the three key architectures based on where the three key functions are located. 

 ALU Small Cell Architecture options

A major new product launch

That's a good point to introduce one of the major new product launches at the show, where Airvana released their One-Cell solution for larger Enterprises. This LTE only system connects remote radio heads using IP/Ethernet to a central 1U sized controller – a DRS system. Airvana have quietly become one of the largest Small Cell vendors, focussing primarily on the North American CDMA market. They've now widened their appeal with this LTE system that is compatible with networks worldwide. It's similar in concept to the SpiderCloud solution except that the baseband processing is handled in the central controller rather than the radio heads. They offer a "toaster rack" option with shared antenna into which multiple remote radio units can be inserted, sharing a common antenna to enable true multi-operator service. They were separately demonstrating a standalone LTE/Wi-Fi Small Cell on the Qualcomm stand (see our video clip below) - this is a different product line and is based on the Qualcomm FSM9900 chipset.

Meanwhile, Spidercloud are now commercially shipping their dual mode 3G/LTE Enterprise RAN.

David Orloff, AT&T, pointed out that most Small Cell deployment today is for coverage rather than capacity, especially in-building. They've developed analysis tools which take several inputs from RF design, social data, transport data and performance data to pinpoint where and when to deploy small cells efficiently. They have already deployed 1000's indoors, in addition to their residential femtocells.

Too many toolboxes and toolkits?

The toolbox approach seems to have become a popular way to deal with several areas which offer a complex range of options. I sensed a little frustration from some quarters that there aren't fewer options to choose from. Perhaps everyone is waiting for AT&T and a few other Tier 1 operators to make those choices and vendor selections first. Another view is that each operator still considers themselves unique with different issues, problems and requirements to everyone else. That makes for good business opportunities for the consultants and analysts who can help navigate them through the available options.

LTE was a lot more prevalent this year

In addition to the Airvana launch of their LTE Enterprise solution, we saw a few new LTE+Wi-Fi products from other vendors. Contela had an LTE+Wi-Fi femtocell based on the Mindspeed (now Intel) chipset. Qucell have ported their LTE small cell onto Qualcomm hardware, although still shipping Broadcom based product commercially. Alcatel-Lucent are clearly also making progress with their multi-mode 3G/LTE/Wi-Fi product (also based on Qualcomm) but weren't publicly demonstrating it quite yet. Mike Schabel, ALU's VP Small Cells, told me the development program was on track for commercial deployment in early 2015. The spec is impressive 4x LTE bands, 2x 3G bands, Dual mode Wi-Fi, Carrier Aggregation, reprogrammable refarming 3G/LTE, Power over Ethernet.

Mike also revealed he had changed his mind since last year and was now proceeding with development of an LTE only femtocell product. His reasoning is that there is a very select set of operators who need very good LTE coverage in order to provide good VoLTE service, liberating 3G spectrum for refarming. There are a handful of important customers who don't have enough LTE spectrum and want access to as much as they can get, buy and reuse.

Overall, it seems that for in-building (residential and enterprise), we are seeing a choice of 3G/LTE/Wi-Fi and LTE/Wi-Fi products coming onto the market. The existing 3G kit is still what's selling in volume today and I doubt if we'll see many new 3G only products in future. The LTE only systems will rely on VoLTE (Voice over LTE) being commercially introduced, targeting specific regions/operators. Delegates from UK operator BT were very much in evidence – they are an MVNO with their own 4G spectrum, and have stated they will use a Small Cell strategy to handle traffic. Many are watching to see how they will proceed – they are known to be trialling many different vendors but keeping quiet on their immediate plans.

WARID, a Pakistani operator, pointed out their country doesn't have any 3G yet and will skip directly to LTE. It's easy to forget that half the world's subscribers are still on 2G, and LTE will be the main method of internet communication. They want a 2G+LTE Small Cell box, something I haven't seen anywhere yet. Orange reported that some suburbs in Eastern Europe consume 30 to 50Gbytes/month via 3G where good wireline internet isn't available – justifying Small Cell deployment.

Learning from experience from the East

Adoption of new technology is often more rapid in some Asian countries. Softbank Japan (who now also own Sprint USA) will soon have 100,000 LTE basestations in their networks. They believe faster, cheaper product comes from a mass produced, multi-vendor market with quality control. Three main areas of learning from their 3G femtocell experience were:

  • BTO: Build to Order. They want separate RF daughter modules which fit onto a common baseband board. This allows them to tailor shipments by frequency band or technology from a set of standard parts.
  • SON: Distributed SON from the same vendor which can run on any vendor's hardware using a standard API. That will please Radisys and Airhop who have integrated their SON capabilities, initially on the Broadcom chipset platform.
  • SMP: Service Management Platform, that combines technical and customer support information. When customers call with problems, the agent can see a complete picture and solve the issue

Others I spoke with suggested that Sprint could achieve a faster pace of Small Cell deployment than AT&T and Verizon with, driven by their new Japanese owners. However, they first have to decommission a number of Nextel cellsites and sort out some major work with integrating Clearwire spectrum/sites into their network.

Presence, location and in-building services

The concept of using Small Cells to track users led to a variety of location based services and solutions. ip.access offer their "presence cell", where a number of standalone Small Cells are deployed not reject rather than carry traffic. In the signalling exchange, the "presence cell" can capture details for use in studying footfall patterns and aggregate traffic flow amongst other applications.

Cisco demonstrated their Enterprise Presence solution, where a standard 3G phone can be assigned an extra office phone number which is only connected when the employee is physically in the office. This doesn't require any download or App in the phone itself or onsite service – it's all done "in the cloud" making it easier to cope with lawful intercept, emergency calls and other regulatory requirements. (Watch a demo in our brief video from the event).

The value of additional Enterprise services is being more widely recognised. Other vendors in this space include Spidercloud who offer their Services Node as part of the Enterprise RAN, and Huawei were also promoting their Service Anchor. Rightly or wrongly, this is perceived as something that DAS vendors don't offer.

Another question arises about who is going to install and deploy these in-building systems. Cisco launched their Enterprise Select program which combines network operators with local Cisco system integrators to deliver Enterprise Small Cells. Cisco partners already have expertise in Enterprise Wi-Fi and data networking, who can be qualified through a Cisco training and certification program. There is strong demand from the enterprise sector for this type of approach.

We've investigated this in depth in our latest White Paper, sponsored by Cisco and iBwave, available to download free here.

A new approach to outdoor Small Cell site sharing

Outside, BT Openreach – the wholesale arm of the UK wireline provider which doesn't own spectrum or directly offer retail cellular services – has signed several agreements with local councils to deploy outdoor Wi-Fi. They've designed a small street pillar that sits next to lamp/light poles with power and wired backhaul. The pillar can be easily upgraded to host Small Cells from multiple operators on a modular basis. Hosting would be offered to all operators on the same commercial terms. With several operators saying that site sharing is the only cost effective way for mass outdoor Small Cell deployment, perhaps this is one way to go.


It was very much a Small Cell event, but the backhaul conference stream was pretty full a fair bit of the time. We saw most of the same vendors that have been visible before, and I didn't spot any really significant new product launches. It seemed to me that many backhaul vendors are in the trial stage. They either need to have other markets that bring in revenue (such as CBNLs recent success in LATAM for Enterprise/Campus fixed services) or investors with deep pockets/patience. Several report successful trials: Arquiva UK with Intracom, Siklu and CCS, BluWan who've replaced a few of the popular Ericsson Mini-link dishes with a much smaller point-to-multipoint footprint. CCS won the Small Cell Forum award in this category, one of many who highlight short deployment time. Siklu promote their automated robotic site alignment tool. John Naylon of CBNL points out that while helpful, we shouldn't lose sight of the big picture and many other factors that contribute to total cost of ownership.

Timing and Sync remain a hot topic, if still a highly technical branch of the industry. Semtech were showing off their USB stick sized boundary/clock, claiming this was being used in older backhaul networks today to make them ready to support LTE-Advanced features next year that require tight phase timing.

Chipset Vendors

I've not heard too much from Intel since they acquired the wireless assets from Mindspeed late last year (including the former Picochip products and staff). Dan Rodriguez told me that these were now fully onboard and that Intel was continuing to invest in the existing (Mindspeed) product, ensuring good technical support and supply chain (which I'd heard had been causing problems last year). This includes their R&D centres in Bath UK, Beijing China and Newport USA.

Although Intel weren't publishing a future roadmap, it's clear their strategy lies more towards network virtualisation and Cloud RAN. They'll be looking at architectures where the power of centralised datacentres can be brought to bear to bring added capacity.

Qualcomm launched a new chip, the FSM90xx, which is a smaller version of their FSM99xx. This is positioned for the residential market, reusing much of its big brother's capabilities and feature set. Their chipset scope is comprehensive and includes the RF front end, GPS receiver, timing, SON etc. It's also 28nm technology and can drive one LTE 20MHz channel in either TD or FDD mode plus dual band Wi-Fi in 802.11n and/or ac modes. This fits with the growing interest in smaller LTE+Wi-Fi only products we reported above.

At the other end of the spectrum (sorry - pun fully intended), Cavium introduced ThunderX, a very powerful 48 core ARM-8 chip, that may be more appropriate for large datacentres and/or Cloud RAN archiectures than traditional Small Cells. It would be like "Using a cannon to swat a fly" as the Spanish say.

Broadcom remain a major force in Small Cells, probably shipping more chips than anyone else today, but keep their future plans quiet. TI position themselves as suitable everywhere except residential femtocells, with a range of chipsets in their Keystone family to suit every performance/price point. Freescale were also present with a stand. Both TI and Freescale are present in many macrocell products, and would be looking to retain a position as the large OEM vendors scale down.

Carrier Wi-Fi vs Amenity Wi-Fi

The Carrier Wi-Fi stream at the conference wasn't as widely supported as the others, perhaps because the WBA is no longer affiliated with the event. Kelly Davis-Felner, VP Marketing at the Wi-Fi alliance, told me that there are over 300 smartphones now certified to run HotSpot 2.0, which should make it seamless and easy to connect to Wi-Fi on the go. Take-up hasn't been anything like as rapid as forecast last year – it seems the Mobile Network operators haven't put roaming agreements in place. Cable operators and WISPs (such as Boingo) are moving faster. Possible reasons include concern about loss of roaming revenue, unknown/unpredictable quality of service and the customer perception that Wi-Fi should be free anyway.

Dave Fraser, CEO of Devicescape, offers an alternative which he believes deals with those concerns and can rapidly differentiate service providers. Their solution builds on existing "Amenity Wi-Fi" provided free by many business owners for the benefit of their guests. They've signed up UK MVNO Virgin Media, and within 2 months captured and validated ("curated" is their term) 50,000 new amenity hotspots which customers can seamlessly sign into. What's really exciting is that their client software is now embedded into Android handsets (he couldn't name the handset vendor but you would know it) which will be in the stores later this month. The client software handles the QoS issue, ensuring good quality Wi-Fi connections and sensibly dealing with handover of streaming apps (i.e. not cutting off your Facetime or YouTube video midstream).


While some frequent visitors to the show felt it was becoming more predictable and there was little new to report, I thought we could see continuing progress on multiple fronts.

  • Another 50% growth in numbers from last year. While ~1000 may be tiny compared to 80,000 at Mobile World Congress, I found I was at least as busy. All the key industry players are here, either exhibiting, presenting or networking.
  • The scope is expanding. Contributions from Enterprise and Local Councils illustrate the wider impact and engagement with other industries. I'd like to see more of that next time too.
  • LTE clearly becoming more significant. LTE+Wi-Fi products will have a market opportunity, it's not only triple mode 3G/LTE/Wi-Fi (athough I think that will be more popular).
  • Carrier Wi-Fi hasn't swept the floor as quickly as might have been thought a year ago. But watch out for some of the other Wi-Fi solutions. Integrated Wi-Fi hardware does now seem to be a mandatory product feature, but not all carriers yet seem to have a fully integrated cellular+Wi-Fi strategy.
  • Scaling up deployment will require partner services and systems. Cisco's Enterprise Select offers up their huge Enterprise partner channel. BT is one wireline operator offering carrier-neutral street level rollout.
  • Presence/Location services seem to be one of the more popular first services to deploy. This will partly be driven by upcoming FCC requirements for better in-building emergency call location details.
  • The Small Cell Forum launched Release Four of their ongoing documentation set, completing the scope of their Urban content. Next up is Rural and Remote for Release 5, which includes not just rural fields but trains, planes, ships, first responder, defence and many other use cases which require a standalone cellular network.

The terminology and scope of Small Cells continues to evolve, bringing opportunities for both confusion and innovation. We'll continue to offer insight, industry views and explanations which bring clarity to this rapidly evolving and substantial market opportunity.

Apologies to those many deserving companies and delegates who we haven't been able to include in this roundup. There's simply too much to pack in, but we'll expand on many of these topics in the coming weeks. So please do ensure you've signed up for our emailed newsletter at the top left of our website. Many leading Small Cell companies actively support these reports and features through sponsorship - contact us to add your business as part of your marketing program.

We compiled a 10 minute video with a few snippets to give an impression of the event, including the new Small Cell Forum CEO, several brief product demos and snaps from the Awards Dinner.

Direct link to youtube video: http://www.youtube.com/watch?v=KQ-IsYrzD58

[Article updated 21 June with minor corrections and clarifications to Airvana's OneCell solution]

Hits : 12554
  • 4




    A significant number of users continue to report poor mobile coverage in their homes. There will always be areas which are uneconomic for mobile operator to reach. They range from rural areas

  • 4




    The term Enterprise addresses any non-residential in-building including hotels, convention centres, transport hubs, offices, hospitals and retail outlets. It's not just intended for businesses to

  • 4




    Urban small cells (sometimes also named metrocells) are compact and discrete mobile phone basestations, unobstrusively located in urban areas. They can be mounted on lampposts, positioned on the

  • 4




    A rural small cell is a low power mobile phone base station designed to bring mobile phone service to small pockets of population in remote rural areas. These could be hamlets, small villages or

Backhaul Timing and Sync Chipsets Wi-Fi LTE TDD Regional

Popular Categories

Follow us on...