RAN trends, antenna evolution and the impact on future site wind load

Right sizing telecom towers for a long-term horizon is tough, there are various factors at play and the industry is evolving fast. Over the past 20-30 years the wind load on towers has been consistently going up but is this trend set to continue or have we hit a peak? As towercos look to plan for longer investment horizons, TowerXchange explores the factors that will influence the future wind load requirements of towers and discusses how towercos can more accurately provision for future capacity.

Over the years, the average wind load on towers has been steadily increasing. New generations of technologies have been overlaid on existing ones and an increased openness towards infrastructure sharing means that more equipment belonging to more people is hanging on towers. The complexity of sites is getting higher and sites in some geographies are becoming overloaded.

Historically towers were owned by the mobile network operators and were not shared. Planning tower capacity for the future involved bilateral discussions between the MNO and their OEM partner, working hand in hand to estimate future wind load and space requirements on sites owned by the same MNO.

The movement of towers into towerco hands has made things more complex. Towercos are dealing with multiple operators on a single site, each with different technologies and rollout plans. Whilst a key partner to MNOs, towercos do not have the same level of access to operator network plans, and have typically not been directly involved in active network equipment discussions between MNOs and the OEMs.

Operators trying to forecast their own requirements for their own towers was challenging; towercos trying to forecast the plans of multiple operators and the impact on towers is even more so. Overlay on top of this transformational changes (5G, virtualisation of the RAN etc.) within the industry and forecasting becomes even tougher.

Yet accurately forecasting wind load is extremely important to towercos when designing new sites or considering upgrades of existing ones. Provisioning over capacity is unnecessarily costly; but so too is provisioning insufficient capacity which necessitates future upgrade or replacement works. Towers are now considered critical national infrastructure, but are generally built to be fit for purpose for the next 5-10 years only. Compare this with other forms of critical infrastructure and you see a stark contrast: roads, railways, electrical infrastructure are all planned to be fit for the next 30-40 years. Towercos need to look longer term and move from a reactive to a proactive state in provisioning capacity.

But all this is easier said than done. Modelling is complex, influenced by a large number of factors which are beyond a towerco’s control and for which there is a lot of uncertainty. Here we explore those variables at play which will influence the future wind load and space requirements of towers.

 

A new generation of technology is, on average, released every ten years. Currently we are in the midst of 5G rollout in developed markets (with 4G still having a way to go in many developing countries). 5G rollout in the macro-layer is involving the deployment of massive MIMO antennae, heavy equipment which is being overlayed on existing towers, adding significant wind load. For this moment in time, legacy equipment is not being removed from sites. Plus, whilst 3G is being sunsetted in a number of markets, 2G and 4G will continue to coexist with 5G for the foreseeable future. According to a recent survey by Analysys Mason, 62% of MNOs expect 5G architectures to increase wind loading significantly by 2024, 68% also expect an increase in space requirements.

This is just what the industry is focussed on deploying now - in around five years we will be talking about 5.5G and by 2030 we will be on to 6G – the evolution never stops! Whilst 6G may be less than ten years away in some markets, little is known about the spectrum bands that we will be working with, making provisioning for the technology extremely challenging.

When rolling out a new generation of technology, oftentimes there can be a particular crunch in wind load, followed by somewhat of a relief shortly after. When the industry moved from 3G to 4G, additional antenna had to be added thus increasing the wind load; the separate antennae have now been widely replaced by a single multi-band antenna covering 2G to 4G, the multi-band antenna reducing the wind load back to what it was pre 4G. The expectation is that there will be some integration between 5G antennae and these multiband antennae, with vendors creating interleaved hybrid solutions to do this. The question is whether the original antenna wind load will increase with this, but the goal is to keep the length and impact on the mast the same.

Equipment vendors are working to create antennae with reduced wind load and space requirements - making antennae more compact or introducing features to improve wind movement across them. MNOs require compact solutions to keep rental fees on towers the same or similar, whilst antennae with a similar form factor keep installation costs down, preventing the need for new brackets for example.

Whilst many vendors have made great strides in improving the wind load of antennae, there are however limitations to how compact they can be. Antenna size is dependent on the number of dipoles, dipole length is determined by wavelength and this cannot be changed. Whilst the new generation of antenna design introduced a couple of years ago enabled antenna widths to shrink significantly, reducing wind load by around 60%, one cannot expect such significant gains ahead. Asking for a more compact antenna is like asking for a litre of water that weighs less than a kilogram – physical constraints mean it is just not possible.

There is a trend towards deeper infrastructure sharing as operators look to control costs in the face of decreasing ARPUs. Whilst towercos had previously seen RANsharing between MNOs as a threat, the new structure of contracts means that it can be turned into an opportunity. Many contracts between MNOs and towercos now provision for active sharing, with higher lease rates payable per physical tenancy if active sharing is involved. One rather than two sets of equipment on the tower reduces wind load, freeing up space for other tenants and reducing the requirement for costly structural upgrades. For now, most RANsharing agreements have been confined to rural areas; in urban areas with high demand, there is less spare capacity to share, and operators still compete based on the quality of their network. To what extent RANsharing is in place will impact future wind load requirements of towers, but the general consensus is that the levels of RANsharing are expected to increase.

Antenna sharing is somewhere between passive infrastructure sharing and RANsharing. Whilst operators had historically been reluctant to share antennae, in some instances they realise it is the only viable solution and they are becoming more open to the concept. From a towerco perspective it is something that they would like to explore more. The benefits of antenna sharing on wind load are clear, but there are some challenges – not least how to integrate two operators using different grids onto one antenna and how to share mMIMO between multiple operators. The adoption of Open RAN technologies will help address some of the challenges, but Open RAN is still in its very early stages and there is a question mark over how widely adopted it will be.

A move towards shared antenna and Open RAN creates an interesting opportunity for towercos – the ability to provide antenna as service. The proposition makes a lot of sense. For a towerco it will give them much more control over what is hanging on a site and will reduce wind load whilst offering additional revenue generation potential. For an MNO it will reduce the amount of capex they need to deploy and will also provide a much more plug and play solution. Many MNO CTOs now come from an IT background and function as CTIOs. The ability for a towerco to take care of the RF element, providing antenna as a service, enabling an MNO to effectively just run the software in a more traditional IT type setting is attractive to them.

Towercos becoming involved in this space also presents a solution to one of the challenges that Open RAN presents. With Open RAN, the idea is to have more optionality in the supply base, enabling companies to pick and choose between different suppliers. The challenge with this is integration and who does it, not many operators have the appetite of Rakuten to recruit 300 engineers to manage the process. If integration proves too complex, we may see MNOs opting for a single supplier to simplify things, thus eroding the benefits of Open RAN. A next generation towerco could include integration amongst its service offerings.

To what extent the industry moves towards Open RAN will impact the future wind load requirements on a tower. If Open RAN progresses well, we will see new hardware vendors coming into existence with new designs which could be smaller and lighter; we are also likely to see more shared antennae, something that will reduce wind load requirements.

The more spectrum, power and frequencies on a site, the more issues with interference between neighbouring cells. PIM (passive intermodulation) issues on sites are increasing heavily as spectrum has become more crowded and infrastructure sharing has become more common. Whilst vendors are working on antenna designs to minimise PIM behaviour, such interference will be a limiting factor on how much new equipment each individual site can handle. Adding more and more antennae and spectrum to a site cannot continue indefinitely.

Similarly, in some countries, EMF issues are also becoming a limiting factor. The sum of all the signals is increasing which will limit how much equipment can be placed on a site as well as the location of the site.

We will see an increasing amount of fibre to the site, with fibre replacing microwave backhaul in some instances. Whilst fibre to the site will increase, the industry does not expect microwave to be phased out entirely. There are a number of reasons to keep microwave backhaul, not least redundancy reasons. Whilst the usage of microwave (and presence of microwave dishes on towers) is set to decrease, the industry expects microwave dishes to contribute to wind load on many towers for years to come.

Obtaining planning permission for towers can sometimes require the deployment of camouflage solutions which can add not only significant cost, but also a significant wind load to sites. The COVID-19 pandemic has to some extent reduced public objection to telecom masts, with the importance of connectivity particularly in rural areas having been brought to light. The industry expects that this might have a positive impact on public acceptance of non-camouflaged site, but in many areas – particularly those of historic significance – camouflage is a must. Whilst this can add to wind load, companies are working on camouflage solutions that help to minimise the impact, with some solutions having the potential to improve wind load.

Whereas MNOs have been the traditional tenants on masts, we are seeing a host of new potential tenants including Fixed Wireless Operators, (FWAs), IoT players and new holders of spectrum (due to flexible spectrum auctions that we are seeing in some markets). For towercos, these present an incremental revenue stream, but they also present additional wind load.

In the case of FWAs, they normally focus on rural areas where sites are not overloaded. The size and wind load of FWA antennae tend to be smaller and so do not currently present a major concern to towercos. Whilst not an acute challenge at present, they do however illustrate how planning wind load requirements is becoming increasingly complex as the type of tenants on sites diversifies.

It is not just new antennae for spectrum holders that are increasing wind loads on towers but also some of the equipment now being hung on towers for different use cases. In urban areas we are seeing lighting, CCTV, advertising and IoT sensors for applications such as weather or traffic monitoring, all of which contribute to wind load. Going forward we can expect further uses of towers which are not yet in use, with drone nests being one idea currently under exploration.

Modelling the future wind load requirements is extremely complex and there are arguably more unknowns than knowns. The good news is that there will always be antenna and towers will always be required, there is no way around the physics of it. But when it comes space and wind load requirements there is no easy answer.

A shift towards deeper infrastructure sharing coupled with challenges presented by large amounts of equipment on sites suggests that the wind load on sites cannot continue to go up indefinitely, even as new generations of technology are rolled out and towers find new use cases. Whether we are close to peak capacity is, however, less clear.

To more accurately provision for wind load capacity over a longer term, there needs to be closer dialogue and better collaboration between MNOs, vendors and towercos. Towercos need to better understand the expected evolution of RAN equipment and need to work closely with MNOs to understand their network plans, finding win-win solutions to rolling out sites more cost effectively. Currently more equipment on a tower equals more revenue for a towerco, but it also means more cost with the building of larger or reinforcing of existing towers. Incentivising both MNOs and towercos to find solutions that reduce the wind load on towers is key to alleviating the growing stress on towers. The antenna as a service proposition is one potential solution, win-win MSAs including RANsharing is another. Virtualisation of the RAN and open architectures have the potential to reimagine the way that networks are rolled out, plus antenna designs will continue to evolve.

The telecoms industry is a fast-moving sector in the midst of transformational changes. Provisioning sites for a 30 to 40-year time horizon with a high degree of accuracy may not be possible. The key to designing sites for the long run is, however, modularity. Towers should be able to be upgraded easily, adding simple building blocks to provision for extra space and wind load requirements. Every time a new generation of technology is added towercos can find themselves removing all the steelwork from a rooftop. The industry needs intelligent plug and play designs that are easy to upgrade without disturbing existing networks.

Towers are now considered critical national infrastructure and whilst there may be no easy answers to how best provision long-term capacity requirements, collaboration across the telecoms sector as a whole will enable networks to be designed, built and upgraded more cost effectively.