The market awaits the arrival of outdoor small cells, but the wait continues. When small cell deployments will happen keeps sliding from one year to the next. I have always thought that some in the industry have been too optimistic on small cells deployment timelines and they have ignored important facts that define the deployment timelines and the shape they will take. In the last couple of months, I spent quite a bit of time looking at various aspects of small cells as part of two market reports that are now published talking in the process to a number of vendors and operators. In “Small Cells New Order” we surveyed leading operators on their plans for small cell deployments. What we found was clear indication that there will be no hockey stick curve yet for small cells. Instead, deployment will be organic and tactical to fill certain performance shortfalls. To put this in context, consider the following point:
1- Operators tested 3G small cells and performance was abysmal. There was simply too much interference from the macro cell to render small cells useless. Furthermore, in some cases small cells caused interference to macro cells and led to negative performance. In short, don’t look for much if any 3G small cell deployments. The window has closed for them even if/where they work.
2- Trials of LTE Release 8 small cells did not demonstrate major capacity benefit, unless they were dropped in the middle of a hot spot that was somewhat protected from the macro cell.
3- Small cells await LTE-Advanced features such as absolute blank subframes and carrier aggregation version of eICIC. Both techniques reduce the macro layer capacity to enable small cells to operate more effectively. Therefore, unless small cells were to be deployed in some quantity, the operators will not recoup loss of capacity at the macro cell.
4- Operators have their hands full migrating to LTE. Even those who deployed LTE are feverishly working to improve LTE performance by refarming 2G & 3G spectrum.
5- The first step in operator’s network capacity improvement is to make the most out of the macro layer. Check out the plans of SK Telecom and EE on this issue – two very different operators in different markets, yet the fundamental economics point to the same path: macro first! What will operators do at the macro layer?: Carrier aggregation (first LTE-Advanced feature to go live – rolled out by a few operators), high-order sectorization (yes, the old 6-sector cell is back in vogue in some markets), higher number of antennas (there has recently been a flurry of interest in 8×4 base stations) are but a few techniques.
6- The cost of small cells is too high. The actual base stations, the backhaul and the process of deploying at low height using light and other poles are just too high in an environment where investors don’t value spending on infrastructure (which in turn led to much CTO-CFO tension at operators that we came across during different engagements). Particularly a problem is securing assets to locate the base stations. Lack of regulation on availability and pricing does not help the small cell business case.
7- Immature SON techniques and wide trust gap will put the brakes on massive small cell deployments. SON is a necessity for network densification (otherwise, it becomes too large and complex to manage cost effectively). Advanced interference management techniques for small cells best work when they are automated – that is they are part of SON. This automation will not take place until operators trust SON algorithms, which will happen after extensive validation. We have not even started this phase (except at a couple of leading operators).
8- Operators are focused on addressing indoor capacity where 70-85% of traffic is concentrated. Wi-Fi offload and DAS are the leading technologies. By offloading indoor traffic, congested macro cells will gain additional traffic headroom.
9- The number of congested sites in a network is relatively small – between 15% – 20%. The statistical distribution is well known (Pareto) and has been proven on many live networks. Small cells will be deployed in the future in such areas, but only when the macro cell capacity roadmap is implemented and indoor traffic is dealt with cost effectively.
The delay in small cells will have profound impact on the system requirements for small cell operations. For example, operators may opt to implement coordinated multipoint with or without eICIC as CoMP solves many of the problems resulting from range imbalance between small cells and macro cells. This will result in new synchronization requirements for small cells and the backhaul link. There will also be increasing requirements on the backhaul link which ultimately I think will favor systems operating in V-band and E-band because their high-capacity and low latency will be key in coordinated access systems.
In “Small Cell New Order” we probe operators on their plans and requirements for small cells and their backhaul. The majority say 2014 is the year for deployment, but the data we have collected actually points that it will be later before they truly start to deploy small cells: in my opinion 2016 when we may start seeing the first serious deployments.