With a view to complementing and creating a similar event to Mobile World Congress, the US wireless industry combined several of their annual events into a single "Super Mobility Week" held in Las Vegas in September. CTIA, CCS (for small rural carriers) and 4GWorld were the highest profile, with the Tower and Small Cell summit being one of the additional attractions. CTIA has had a poor reputation in the past; 4GWorld has also had a declining reputation – could this new approach win back support and make it a credible counterpoint to MWC?
A regional event
Undoubtedly, this is a regional event for North America rather than a global event. Keynote speakers proudly announced that the US is far ahead with 4G:
47% of all global LTE subscribers are in the US, despite only comprising 5% of all global subscribers and being allocated the 3rd lowest amount of LTE spectrum. The headline data rates are impressive (typically 30% faster than in Europe) and 4G provides a platform which accelerates App development and innovation. So perhaps it would be churlish of me to point out it doesn't yet offer a 4G voice service. Sprint's CTO said they wouldn't rush in to launch VoLTE until all the technical issues were resolved, and instead offer HD Voice on 3G CDMA.
The keynote speeches were frankly a bit dull – CTIA insisting on more spectrum and less regulation for what they claim is an industry that continues to power a substantial part of the economy. As an outsider, I'd point out that Americans also enjoy some of the highest prices for their cellular service and wonder if the radical marketing initiatives by T-Mobile and Sprint will make a substantial impact as we've seen in France and Japan. If cellular service is so great here, why is the industry shipping over 100K Femtocells and repeaters every month to fix poor residential coverage problems? To address the growing numbers of smartphones, tablets and other gadgets, there's growing popularity with "family plans" providing a single bill for multiple devices. Examples are Verizon offering 4 devices with unlimited talk/text and 10GBytes for $160; T-Mobile offer the same for $100 or with unlimited data for $160.
By comparison, my own UK provider offers free roaming to the USA, so my humble $15 plan allowed me to use my 600 minutes and 1GByte data package at no extra charge.
Diverse and somewhat unbalanced participation
From a Small Cell perspective, there were relatively few industry representatives. DAS vendors were here in force, leading me to think the side event should be titled "Tower and DAS summit" to reflect its composition. While many would now include DAS within the scope of Small Cells (a typical definition is "anything that's not a macrocell"), the workshop panel sessions were somewhat unbalanced and not hugely informative or confrontational. But the two day event left little time to fit in enough meetings, and from which I gained some excellent insights. The overall impression is of an event that tries to appeal to all, covering all aspects of wireless, but reaches critical mass in few. More than one attendee commented that they found smaller targeted events more useful.
Ericsson wakes up to Small Cells at last
Biggest news announcement of the show for me was Ericsson's new RBS6402 small cell product. After years of saying they're not needed, and making a lot of noise with their Radio DOT, they've come up with a pretty good box. It's aimed at the small to medium Enterprise and provides two bands which can be both 3G, both 4G or one of each. 4G carrier aggregation is also possible. It's software configurable to any of about 10 RF frequency bands. Powered by Ethernet, it will handle up to 128 4G and 64 3G concurrent active users. Wi-Fi at 2.4 and 5GHz is also included. They claim a 10 minute commissioning time. Ericsson don't have an LTE gateway at this time – it connects directly into the LTE core, so it may not scale as well as other vendors but this is something that could be addressed later. Based on the same software as their largest cousins, this is sure to be of interest to existing Ericsson customers. I was unable to obtain a full datasheet or confirm the date of commercial availability.
While this is positioned to work best underneath Ericsson macrocells, the key factor will be pricing. I think they've held off launching this product to protect their macrocell equipment sales which will suffer if this is to be competitive with other small cell vendors. They confirm the potential market is huge, stating that over 10 million buildings of less than 50,000 sq. feet need this product, in addition to larger buildings where they would target their Radio DOT solution. Video interviews with Ericsson CMO from RCR wireless here and here.
The Billion Dollar DAS market
I'd estimate that the global market for DAS (Distributed Antenna Systems) today is about $1 billion/year, excluding the basestations that power them. It's not uncommon for an installation to cost $1m or more. US Federal regulations demand that large building owners provide coverage in case of emergencies. Equally, building owners want to service their guests with both cellular and Wi-Fi connections.
Perhaps to illustrate this, my phone made a loud noise as I received my first emergency text alert, warning me of torrential rain flooding the local streets.
DAS system design varies significantly – they all basically take in the RF signal from multiple operator basestations, combine them and pump out the signal through a series of remote radio heads. Some transmit the RF signal purely as analogue, others do this digitally and even interface with the basestations via digital CPRI connections. Some products are more suited for the highest capacity scenarios (e.g. football stadiums) while simpler/cheaper options may suit lower density office environments better.
Boingo, who I had thought were purely a Carrier Wi-Fi aggregator, explained that 25% of their business is now from DAS (i.e. $25M/year). They offer a joint neutral DAS and Wi-Fi service to property owners – saying that the combination of both, all working across different carriers and technologies has been a compelling one. They have Wi-Fi roaming agreements with 3 out of 4 US major networks and DAS that works with everybody.
Tower business evolves to street level and in-building
There were quite a few Tower installer businesses there, considering how to evolve and accommodate the new small cell rollout scenarios. It won't be nearly as lucrative a business in the longer term. Indoor/enterprise is more likely to be done by existing IT teams; outdoor street installations align better with Cable TV installers. Small cells today remain a relatively small revenue business today compared to the $1 billion DAS or American Tower's $3 billion turnover, but are forecast to grow by 65% to $1.3 Billion this year.
Panel session discussions revealed that it's taking 5 or 6 site visits to install each street sited small cell today. The target is more like 1 or 2. Partly that's down to some of the macrocell deployment practices that just don't translate to high volume. The major networks are struggling to learn and perfect these procedures. Shaun Cohen of Vertix Consulting told me that several of the large Cable TV companies he works with are positioning themselves to be able to hang small cells from their existing strands (wires suspended in the streets). They should be able to do this at high volume/speed using their existing field staff, processes and assets.
For me, Tom Wheeler from the FCC delivered the most interesting speech. His three key points were:
- Competition delivers stronger innovation than regulation. They've been outspoken to prevent consolidation, blocking the acquisition of Sprint and T-Mobile, believing consumers benefit from the renewed competition.
- Net neutrality: He sees an incompatibility between "unlimited internet" and bandwidth throttling – doesn't agree can credibly market both together. They want to treat cellular on the same basis as fixed broadband, something the CTIA strongly argues against.
- Spectrum: The CTIA lobbies hard for more spectrum. Broadcasters believe demand isn't as strong as stated. So the government will auction off broadcasting spectrum and let the market decide which parties see it as most valuable.
A couple of selected comments from other executives:
- Michel Coombes, CEO Alcatel-Lucent: Small Cells are a clear new frontier for our industry
- Sprint: Small Cells are a big part of every carrier's plan to make the best use of spectrum. We have some challenges in deployment scenarios, finding cheaper street furniture sites, suitable backhaul – these challenges haven't gone away but will be resolved.
Various exhibitors and announcements
Samsung report that they are shipping increased volumes of their 3G CDMA femtocell. Their major customer is Verizon who had announced Samsung as a supplier for 4G LTE Enterprise in February. This will be a 64-user LTE only product and important to support the rollout of VoLTE in-building. Commercial product launch is expected soon, but no firm date yet. Sprint had also previously announced selection of Samsung small cells for 4G.
The major RAN vendors were present here as you'd expect: Ericsson, Alcatel-Lucent, Huawei, Nokia, Samsung... but very few independent small cell vendors.
Fibre backhaul everywhere?
Towercloud, who provide fibre backhaul end-to-end in many south-eastern US markets, strongly believe that fibre would be used in the majority of outdoor small cell deployments. They see macrocell sites shifting from 80/20 fibre/wireless backhaul to more like 95/5 in metropolitan areas. They've connected 22 small cell sites at Altanta Olympic Park for Verizon, each with 4 new dark fibres. While that seems overkill to me, you can't argue it provides plenty of room for growth.
Others commented that the situation varies between countries and cities. While fibre is easily and cheaply available in Japan and Korea, this isn't the case everywhere. Fibre will remain the preferred choice and become more widespread, but the faster speed, flexibility of wireless links for the "last mile" on the street is more common. Mainstream wireless backhaul suppliers were evident throughout the show, with a few of the newer/smaller companies visible or attending. Vubiq were showing off their 60GHz product (no longer includes a built-in alignment motor) which launches commercially in 6 weeks. Fastback, Tarana and a few Non-line-of-sight vendors were there. Most comment that the volumes for small cell backhaul aren't yet sufficient to warrant heavy marketing spend, and look for other market segments to carry their business forward until it picks up.
A significant market for repeaters
Many countries ban repeaters unless supplied and controlled by the network operator. The US does allow consumers to buy their own repeaters independently, and these can be multi-band and multi-operator. The largest company in this business, Wilson, who recently acquired zBoost, told me they ship about 40,000 units a month - mostly they are priced in the $300-$500 range. Their products support multiple bands and can handle both 3G and LTE. Nextivity demonstrated their 3rd generation product which can be remotely controlled if needed and opens up a host of configurable options for professional installers to tweak. In both cases, the idea is for these devices are self-managing and suitable for self installation. It's estimated that there are still more femtocells shipping than repeaters, perhaps because repeaters don't add extra capacity. They're clearly useful for remote areas without wired broadband and low signal strength. Surecall were also active at the show - I didn't realise their high end repeater products could cost as much as $3,500 retail.
KUMU – a game changer to watch out for
We reported on KUMU at Mobile World Congress. These clever Stanford University Professors and post-graduates claim to have defied the laws of physics and developed a system that can transmit and receive on the same spectrum at the same time. "There is no DD anymore – not FDD or TDD". Steven Hong told me that they've come up with a quick way to implement this technology with standard basestations, connecting a supersmart echo-canceller in place of the usual RF rx/tx combiner/splitter.
One application is to use the spare macrocell capacity that this creates with nearby small cells, effectively replacing the need for extra backhaul. The small cells become more like relays, using the same spectrum to deliver the service as for backhaul. A key advantage is that no special handsets or macrocell basestations are required. The small cells would be standard and use a special receiver for backhaul. It all sounds far too good to be true, but I'm told that initial proof of concept trials with four Tier 1 operators have been successful and are now moving on to the next phase. Field trials using equipment from major RAN vendors are scheduled for 2015 with commercial product deployment by 2016.
The solution would allow the full 100Mbps capacity of each LTE macrocell sector to be spread between 3 or 4 nearby small cells, which would only require power at each site.
The technology also has applications for combining LTE-U and Wi-Fi in the same box (Access Point), resolving aspects of co-channel interference.