Progress with urban small cells seems to have been much slower than originally forecast. Has this market died, gone to sleep or is it just resting? For another perspective on this, I caught up with Arthur Giftakis, SVP Engineering and Operations at Towerstream, a leading Fixed Wireless Fiber Alternative provider, offering shared wireless infrastructure and backhaul solutions in top urban U.S. markets.
Any visible signs of progress?
Contracts have now been awarded to our partners, the major network equipment vendors, for urban small cell products. We would like these to be for thousands at a time; we aren’t at that level yet, but can see growth. Deployment is starting - on light poles, telephone poles and rooftops throughout major cities and boroughs.
After extensive trials and tests, the major carriers have settled in on what their small cell specifications should look like and are now starting to execute. Compared with the maturity of macrocell rollout, it has taken time to learn just how much is involved in the procedure. For each small cell, every rooftop or light pole site is unique. This all adds to the deployment process and the time required has taken some by surprise.
Unforeseen delays have arisen for many reasons. It’s taken longer to ratify the equipment, agree on the financial model, determine what makes sense commercially and ensure how it fits with macrocells.
There have also been many distractions such as new spectrum auctions, M&A activity, unlicensed spectrum (LTE-U) and Cloud RAN architectures.
Network operators have some pretty bright people working in their strategy departments who constantly evaluate and assess all these new technologies. If a good business case can be made to adopt a particular approach, then they will take advantage of it.
LTE-U for outdoor small cells? I thought this was intended for indoor only.
Perhaps surprisingly, the 5GHz band isn’t used very much outdoors at the moment, so doesn’t yet suffer from the high congestion levels sometimes found indoors. We have over 3,000 outdoor Wi-Fi access points in our network and have had great performance with the 5Ghz band. The physical characteristics of 5Ghz lends itself to be very attractive for small cell outdoor deployment.
Making more use of 5GHz outdoors depends on your perspective. Unlicensed spectrum is commercially very attractive. Qualcomm and Verizon are strongly promoting the use of LTE-U. Although it’s not yet standardised, it could be rolled out across the U.S. commercially pre-standard. I think it will make a play – the technology has legs – although there are a few important aspects still to be worked out. It must be designed to play nicely rather than overwhelm Wi-Fi. Most smartphones should be capable of being adapted to support LTE-U in future versions – it’s not a huge technology leap from today.
At this stage though, I don’t believe operators will wait for LTE-U products to become commercially available before rolling out small cells. If products had the potential of being firmware upgraded to support it at a later stage, then that would be a bonus but isn’t mandatory.
A couple of years ago I would have definitely said that operators would deploy their own carrier grade Wi-Fi integrated into urban small cell equipment on a mass scale. Now I’m not so sure about operators deploying their own Wi-Fi outside specific high profile venues and areas or mandating their Small Cell equipment has Wi-Fi built in. I think we can expect to continue to see wholesale agreements with aggregators and larger Wi-Fi players such as iPass, Boingo and Towerstream of course, to facilitate traffic offload. That would mean small cells would only need to be cellular only, possibly supporting LTE-U.
Is the term Urban Small Cell still well understood?
What the industry means by “Small Cell” is getting very watered down. It’s almost come to mean anything below a full macrocell. For us, this isn’t causing confusion or uncertainty – densification still means there will be a need for equipment at many more sites and we don’t really care too much exactly what that is. Fundamentally, the appetite for data traffic continues to increase where spectrum is finite, so it’s inevitable that we must have densification.
All architectures have their own merits and drawbacks and I’d expect to see a variety based on factors including backhaul and spectrum availability.
Are the US operators visibly taking different strategies?
We talk to network operators weekly, and although I can’t speak for them directly, I can share a couple of examples of what we’re seeing and provide my own interpretation.
Verizon appears to be favouring a Cloud RAN approach, relying on dedicated dark fibre to every site. That’s not always feasible – try installing that to the third floor of a typical Brooklyn building – the economics just wouldn’t make sense. They also have extensive spectrum assets and so would concentrate first on maximising full use of existing macrocell sites with that before needing to densify through small cells. They have publicly set aside $500 million for small cell deployments focussing on where they didn’t win AWS spectrum. Publicly quoted numbers are fairly low with 400 planned for San Francisco to cater for the upcoming Superbowl next February.
Sprint on the other hand have announced that contracts have been awarded for 70,000 new outdoor small cell sites. This may be influenced by their Japanese owners, Softbank, who have extensively adopted this approach. That figure includes many pole mounted sites rather than just rooftop or wall mounted ones, which will accelerate rollout speeds.
For all networks, the focus on capacity growth outside is to use LTE rather than 3G.