Will 5G power provisioning be the defining moment for the tower industry?

At any given time, the energy from the sun that strikes the Earth is more than the world’s total energy use – more specifically, earth receives 173,000 terawatts of energy from the sun per hour, sufficient to light up the entire world for a period of a year.

 

However, the constant gap between receiving such energy and harnessing it has always been the efficient and effective storage of energy, and the telco industry is no exception. This pain point has become even more prominent with the increased commercial rollout of 5G networks.

 

The telco industry consumes around 2-3% of the world’s total electricity produced – that translates to approx. 4,833 terawatt per hour per annum. To provide some context, such power is consumed by a combined 7 billion mobile devices, 6 million telco towers with 1.5x co-location, 100% 2G and 70% 4G systems in the towers. And the target is to have up to 10 million telco towers with a 70% 5G conversion within a decade from now (figure 1).

Traditional 2G and 4G systems consume approximately 4-6 kW of power per tenant and 2G+4G+5G consumes approximately 12kW per tenant. Addressing this exponential growth in energy requirements for 5G sustainably is a major challenge faced by the telecommunication service providers, and the energy capacity requirement is expected to increase as 5G maturity rollout intensifies (figure 2)

The evolution from 2G & 3G to 4G saw a 43% increase in power requirements. However, 5G anticipates a further 68% increase in power capacity needed from 4G, at a consumption rate of ~11.5kW. An internal combustion engine would need to drive 500,000 km to consume the same power needed to run a 5G site for one year. Considering that an average car travels ~23,000km per year, a 5G site’s power consumption more than doubles that of an average car per 1 unit of time. 

Currently, network technologies leading up to the 4th generation of mobile networks (1G, 2G, 3G & 4G) mainly use low frequencies not exceeding 2.5GHz. 5G however, operates on a combination of mid-to-high frequencies ranging up to 52GHz. These high frequencies enable the transition of theoretical network speeds from 1 Gbps for 4G up to 10Gbps in 5G. Use of mid-high frequencies could deliver higher data speeds, more capacity and low latency, but it also requires more power to generate.

The waves also lose their energy quickly, being able to travel only about 2% of the range of 4G. 5G waves are very susceptible to physical barriers as well. This denotes that it requires more energy to generate as well as needing considerably more small cell sites to eliminate any dark spots, leading to very high-power requirements.

That said, 5G has the potential in the long term to be more energy efficient with idling techniques. Plus, while more energy is required for 5G mmWave, they also carry exponentially more data. 5G adoption is still at its infancy as of today, hence the full potential of 5G has not been realised yet. However, when 5G usage is more widespread, the energy cost per transmitted GB has the potential to reduce, due to increased scale and capability of being able to transmit to, and with the consumption of higher capacities.

As shown above, the higher the transmittable capacity per second, the lower the cost per 1 unit of capacity, even with 68% higher energy requirement.

 The cost of energy for 5G will be high for both the telecom and tower industry, especially in emerging markets where grid availabilities may vary drastically depending on location. Lower grid availability means the need for power back-up systems, which will increase cost to the parties involved. The main challenge of 5G is the uptime sensitivity and with the 5-nines uptime requirement, even good grid sites may need to be equipped with some form of back-up solution.

However, conversion of sites from 99.99% uptime in 4G world (52.6 minutes outage p.a.) to 99.999% uptime in 5G world (5.26 minutes outage p.a.) for a scattered access site network would be challenging without ensuring the actions below.

Mismatch of unit economics cost vs revenue formula in the towerco world?

The energy charging formula used by the tower industry to telecom industry is indeed a decade long one. It is adjustable for few variables such as grid/ diesel consumables as well as power load. Both the items are adjusted in a linear formula. However, the actual cost (assuming the total cost of ownership is 10 years) is exponential. As a result, there will be a point in which the cost of power will increase exponentially for towercos. Figure 6 shows a sample exhibit of the cost recovery line for towercos in the current 4G era and the expected exponential growth in investments needed for deploying 5G compared with the current charging mechanism.

The most immediate solution towerco could pursue is to revisit the charging formula for 5G to reflect the non-linear cost increase. However, the tower industry could also consider the following long-term solutions to reduce the cost burden:

Every penny lost in the power system is a loss to the industry. Each power element ie, battery backup, generator, rectifier should be equipped with sensors and advanced analytics to detect and rectify any leakages and drops in performance. Although it may seem counteractive to invest more capex in order to reduce future costs, this solution would undoubtedly unlock future savings, especially when the towerco business model is a generally long-term business. 

Even with precise network planning, sites are often built at locations with a lack of commercial viability, and such insights would only be visible to the MNOs long after the site has been erected.  This is even more relevant to 5G cells where the network traffic is still unpredictable due to a lack of historical data and 5G mmWaves being the most susceptible to physical barriers. A solution could potentially be the usage of smaller lamp poles instead of macro towers for 5G deployment, allowing sites to be more mobile and miniaturised.

Although the global tower tenancy ratio is around 1.5x, the co-location of power solutions at these sites is only at an estimated 1.0x, and this is despite independent towercos being the preferred rollout partner. This means there is hardly any consolidation of power systems, something the industry can’t afford in the 5G era.While each MNO has its own unique power requirements, towercos could facilitate the consolidation and sharing of power solutions on site. Towercos are able to assess and distribute power more efficiently to each tenant on a site compared to an unconsolidated model. 

Traditional RAN architecture dictates that the access network portion has to be from the same vendor in order to interoperate. OpenRAN architecture challenges the norm by enabling hardware and software components to be modular and interoperable between different vendors. This creates an opportunity for single RAN consolidation which reduces the amount of equipment needed to provide their an MNO’s desired level of service. Some MNOs will leverage the lower deployment costs, while some enjoy multi-Radio Access Technology (RAT) capabilities that enable them to renew and refresh their networks (adding new 4G and 5G services on existing equipment and infrastructure) while maintaining legacy services.

Commercial power providers implement a demand charge during peak usage hours. To mitigate this, intelligent power storage systems would charge batteries during off-peak hours in order to then run-on batteries during peak hours. The systems will insightfully determine the schedule for the batteries to be charged and discharged, saving capacity for power outages.

In any case, 5G power will undoubtedly be a defining moment for the overall telecom industry and more specifically for the towerco community, challenging and redefining the current business framework as well as the operating model. The industry needs to pay a close attention and perhaps even improvise things, in order to make the 5G business case positive for investor groups, end users and infrastructure suppliers in the industry.

Gayan Koralage is one of the pioneering members of edotco Group. He speaks and writes frequently as a thought leader in the mobile and neutral party host telecom tower industry, covering key topics of business case for 5G, network disruption, digital economy, digital transformation and inter-generation opportunities in the 2020 decade. Gayan currently serves as the Director, Group Strategy, responsible for long term strategy, pricing and commercials, analytics, bite sized M&A deals, new market entrance performance analysis and project management. And served additional interim roles of acting country managing director of Pakistan and Sr Lanka. He spearheaded the formation and growth of edotco since 2013, and the sale of 32% stake of edotco to Japanese and Malaysian sovereign wealth funds in early 2017. 

Gayan joined Telekom Malaysia International (“TMI”) in 2005, where he was the Assistant Vice President of Group Corporate Finance, Mergers and Acquisitions and Special Projects. 

His credentials span financing, funding and investment initiatives across the TMI footprint, business turnarounds, running end-to-end processes and negotiations in mergers and acquisitions for Axiata Group, as well as post-merger integration initiatives.

He has been instrumental in finalising key M&A/ Corporate Finance/ Business Development projects for Axiata Sdn. Bhd. in Bangladesh, Indonesia, Sri Lanka, Pakistan, Thailand, Cambodia, Malaysia, Iran, Myanmar, Laos, East Timor, Maldives and Philippines. He has been significantly involved in corporate exercises involving Axiata’s subsidiaries.Gayan is a Chartered Accountant in the UK (CIMA) and is a Graduate in BSC accountancy and Financial Management from University of Sri Jayawardenapura, Sri Lanka.

e

.