A sunny forecast for the future of infrastructure sharing

Cellnex’s Oscar Pallarols, Director of Innovation and Product Strategy, predicts that increases in both connectivity and mobility will create a huge opportunity for the right players to offer infrastructure services well beyond simple ‘grass and steel’ models. We discussed Cellnex’s plans for capturing this market, timescales for realistic rollout and the factors which will see macro infrastructure remain relevant for years to come, albeit with new models for sharing and ownership.

TowerXchange: Let’s start with the current situation - what are the main drivers which have got European infrastructure to where it is today?

MNOs trying to spin off their tower assets as they look for growth and to free up cash flows is not something new, it’s been happening for some time, but it’s an important part of the tower industry today. There’s a flexibility in Europe in that they can spin out their infrastructure, sell it to a third party, or even create a joint venture sharing infrastructure with another MNO - network infrastructure demands high investment and there are many ways to mitigate this.

I think rural coverage is another key factor in rollout to date. Obligations for mobile broadband coverage are harder to meet when looking at very rural spaces. I don’t see it as a driving force for towerco growth, it’s largely driven by government funding in order to invest in rural areas and provide what’s meant to be a universal service. Elements of this still need to be clarified and what can really be shared needs to be identified.

Globally speaking I see two big trends. The first is operational excellence. The P&L of any tower portfolio is looking at energy efficiency, operations and maintenance, tracking the sites remotely as much as possible and obtaining (and using) data to address problems, not just reactively but proactively and preventatively. The second trend is new business models. This is related to the sharing and management of active equipment. Towercos are looking increasingly to small cells and DAS; it’s not consolidated yet but it will become an important part of growth for any towercos in the next future.. We keep a proactive approach on this and are taking strong steps in this direction.

TowerXchange: What do you see as Cellnex’s strengths in the current market?

I think one of the first things is our investment capacity. We are willing to increase our footprint but we also have a huge plan to reduce overlapping by rationalising sites, and become more efficient from a resource point of view. We’re also evolving beyond the traditional towerco business as we’re taking a very strong position in the DAS and small cell world. In terms of roll out, we think it will take longer than predicted as MNOs try to leverage as much as possible from their existing macro infrastructure based on  technologies like M-MIMO beamforming, which can multiply data throughput by 4-5 times thanks to spectrum efficiency; or LTE Carrier Aggregation that allows to combine multiple carriers with maximums of throughput of 600Mbps; as well as Massive IoT, which allows coverage improvements, with for example +50k simultaneous connections per base station. All of these technologies or a combination of them will allow them to squeeze more from the current deployed infrastructure.

Small cells are efficient for crowded areas and in-building solutions, but they’re part of the marketing for a new wave of technology, and we’re hearing huge numbers, talk of smart cities and new deployments. That doesn’t mean the hype isn’t true, but we expect rollout to take longer than predicted.

We’re a company with the investment capacity to grow and expand our European footprint, and that allows us to work with the same customers across boundaries, and in several countries we can create common strategies. This cross border experience will be a big driver for growth in the coming months and years.

Densification is another growth vector, although it will massively happen from 2020 onwards.  Most probably this will require and determine a higher sharing ratio. It does not seem realistic that each MNO will be deploying their own proprietary infrastructure as they did with macro cells. The sharing concept will have a much more prominent role in the market, and there will be a role for neutral infrastructure operators.

Thus, we believe that rationalisation and densification will be driving forces for Europe. We’re leveraging our scale and interest from vendors to start creating a new products and services roadmap. OEMs such as Ericsson, Huawei and Nokia want to team up  to test new operational procedures and equipment, which is working well for all parties.

TowerXchange: What does Cellnex believe the timescale will be for 5G rollout, which markets will embrace it first and what will drive it?

I think we should make sure we’re differentiating between 5G deployment in the largest cities in certain countries and 5G deployment at scale. That will be determined by really interesting use cases from the 5G point of view which don’t originate from mobile; things such as health, vehicles, augmented reality or virtual reality. As yet there aren’t any robust use cases massive enough to foresee when a 5G deployment at scale will happen. One of the sectors which is looking most likely to drive this forward is the automobile industry.  Autonomous vehicles will demand resilient connectivity able to guarantee massive coverage without white spaces as well as latency and suitable speed ratios.

Video will also be a big use case as it needs a lot of bandwidth. Sure, it will happen but I have doubts about the rollout timeline.. There is nowhere currently with a clear plan for 5G involving a set of shareholders like MNOs, towercos, et cetera. As such there is no reference from which you can understand what will happen worldwide. There are lab tests to validate low latency or bandwidth but there’s nothing which has been proven on a functional scale.

If you look at the investment required, there are four big pieces. Number one is fibre; if you look at Spain there’s a lot of fibre connectivity. Telefonica, for example, have invested heavily in fibre in the country, Spanish fibre addresses 70% of the population already, whereas in the UK it’s less than 50% of the population. Secondly, we need to look at what the spectrum fees will be in different countries and if the amount of spectrum increases will the price we’re paying for it remain the same? Thirdly, in terms of the price of equipment, if it’s all shared there’ll be a lower TCO as one box will serve three or more operators but it needs to be complemented with a fully operational platform. Finally, the investment to provide the service to 5G is based on use cases which have not yet been identified.

TowerXchange: Talk us through how Cellnex is preparing for 5G rollout and how your existing competencies can support that.

We acquired Commscon 18 months ago, a small cell and DAS company which was a functioning business we could build on top of. We recently leveraged their expertise to deploy a DAS network in the Atletico de Madrid Stadium as a neutral host. The acquisition of Alticom in The Netherlands brought us broadcasting as well as telecom services provisioned from high towers, but the reason we liked the asset was because their know how on data centre location as well as edge computing as this will play a critical role in keeping latencies lower in 5G infrastructure. Thus, storage and computation will move closer to end-users to enable new services and applications that will live and grow within the 5G ecosystem.

We have other work streams in the pre stages, plus innovation and research activities. One example would the “5GCity” project, a 2020 Horizon Programme project funded by the EU. . It involves three major cities - Barcelona, Luca and Bristol - aiming to deploy a distributed cloud and radio platform for 5G neutral hosts.

In terms of multi operator and multi band small cells, we will be managing the site as a service and based on the demands of the MNO. We’re testing this out at the moment but we have a good asset in house as we manage end to end networking in Barcelona already, and we have the expertise to manage the network from one end to another (with the exception of owning the license). We have the engineers, control rooms and the understanding of different technologies.

TowerXchange: Looking ahead to five years from now, what do you envisage the major changes in communications infrastructure being? Will things have changed dramatically?

There will definitely be changes. Densification will happen. For instance, if we talk about the Mediterranean coastal resorts in France, Italy, Spain and Greece, tourism will be a driver to deploy a second level network for capacity hotspots. Secondly, there will be more sharing scenarios as services will converge, reducing investment in opex. There will also be a change to the towerco’s need for fibre, all the traffic needs to have high speeds at the core of the network which will only happen with fibre rollout as it’s much more complicated.

TowerXchange: Skipping ahead to 2030, what is Cellnex’s vision for your role in the market? Do you think ‘towercos’ as they exist today will have a role to play? What competencies will be needed?

The role of towercos in 2030 is hard to assess. Maybe in 13 years all this will be more or less a commodity, MNOs will converge on third party networks and maybe also spectrum. They’ll become more focused on content and solutions for specific verticals and use cases. There will be a huge amount of devices connected, the popular figure is that there will be 50bn people and things connected in 2020 - we have no doubt the figure will be hit, but perhaps not as soon as 2020. The level of connectivity and mobile connectivity will increase dramatically - both the amount of connectivity and how many moving parts there are. Networks will be more dense and infrastructure will need need to be more dense. Macro cells (if they still exist) will be shared more, whoever manages and operates them - we can be sure that sharing ratios will increase.