Will 5G facilitate or hinder our quest for a carbon-neutral future?

The introduction of 5G networks will act on global carbon emissions in 4 domains:

• Energy consumption of telco networks
• Energy consumption of devices
• "De-carbonisation" of energy supply
• Reducing energy demand from enterprises and consumers

On the face of it, 5G promises to be a key enabler in our quest for a carbon-neutral future. In simple terms, due to much lower kWh/TB transmitted, 5G should outperform 4G by an order of magnitude and 2/3G networks by many orders of magnitude. Mathematically, carrying more traffic on 5G should translate into lower energy consumption than leaving it on 4/3/2G networks. The potential catch is that 5G will accelerate volumes and energy consumption from new applications (such as more immersive gaming) and/or displace traffic from even lower energy fixed or WiFi connections. In some respects, this mirrors the challenge faced by the airline industry which is constantly reducing emissions per passenger mile traveled but also seeing an inexorable growth in volumes. The big difference however, is that the airline industry is not enabling rapid transformation in other industries. 5G has a key role to play in what STL Partners has termed the Coordination Age.

We are modelling this as part of a wider research project and will be publishing findings in the coming months. More needs to be done, particularly to inform policy makers and telcos themselves in how to ensure that 5G can indeed enable a carbon-neutral future sooner.

Energy consumption of telco networks

MNOs' main direct carbon footprint originates from their networks with around 2/3 of total telco energy from access networks. Telcos can contribute to reducing their own direct carbon emissions by reducing their energy consumption and consuming lower-carbon forms of energy (although here, they largely rely on others' efforts in de-carbonising electricity generation). With traffic volumes increasing 30-40% annually, simply maintaining energy consumption levels is challenging. Reducing these is even more so. This will not be possible without moving to 5G for the reasons set out above: due to much lower kWh/TB transmitted, 5G should outperform 4G by an order of magnitude and 2/3G networks by many orders of magnitude. Mathematically, carrying more traffic on 5G should translate into lower energy consumption. Determining how much lower 5G's kWh/TB will be is challenging (not least due to practical considerations, such as availability of sites) and we will need to see the results from actual deployments.

Energy consumption of devices

The second area where 5G has a potential impact on global carbon emissions is in the energy consumption in devices. In practice, a number of factors (including device antenna designs) contribute to device energy consumption. For smartphones (by far the device category with the largest contribution to carbon emissions), 5G's higher frequencies and NR will result in lower device energy consumption for the same applications and throughput. In practice, we can expect 5G to result in more energy and throughput-intensive applications being enjoyed for longer periods. However, battery technology is only seeing modest annual improvements and will therefore continue to constrain energy consumption. Also, gamers who power their devices in will find them too hot to handle for long sessions.

Other devices (especially the billions of sensors and actuators that we anticipate) will benefit (in terms of energy consumption) from optimised 5G slicing although here the alternative may not be similar 4G devices consuming more energy but no devices at all.

"De-carbonisation" of energy supply

As we move to greater reliance of renewable energy to supply our homes, transport, offices and industries we will need to get much smarter with how we manage the supply volatility that is inherent in renewable sources such as wind and sun. Partly, this can be addressed through demand control (e.g. washing machine turning on when supply is plentiful) and distributed energy storage (e.g. treating electric vehicle fleet as an enormous battery). This means getting much smarter in predicting and responding to the balance of supply and demand. It also means squeezing more out of renewable sources. This will require reliable, secure, inexpensive connections to billions of sensors and actuators. Although other networks may well be able to serve part of this need, 5G was largely conceived with these applications in mind and will do a far better job of it.

Reducing energy demand

The largest potential impact from 5G and the hardest to estimate will come from reducing energy demand by reducing inefficiencies and waste from (public and private) enterprises and consumers. This will be an indirect impact in that enterprises and consumers will reduce their energy consumption and carbon footprints by adopting applications that are (more viably) enabled by 5G technologies.

These applications will not necessarily have energy reduction as their main objective. They may be primarily concerned with optimising other resources (e.g. machines), saving time or reducing non-energy costs (e.g. maintenance). For example, an augmented reality application supporting field workers and data from on-site sensors will reduce the number field visits, which in-turn reduces both total distances traveled annually by field workers and the number of vehicles (and field workers) needed. For consumers (and prosumers), 5G will grow the sharing economy by enabling remote safe access to homes, cars, rides, pets, or food that is approaching its "sell-by" date. Reduced energy consumption will be a considerable, if unintentional benefit from these 5G-enabled applications.