Two people working on a laptop.

Shaping the next generation of mobile communications

From the Internet of Things to smart cities, the move to 5G technology promises to have a huge impact on the world around us, bringing internet-connected devices into everyday life.

But before investing in new networks, 5G providers want to be sure that the infrastructure is as robust as necessary to meet users’ needs. Not only that, as their investment could run into billions of pounds, they want to prevent any unnecessary overspend too.

That means understanding how to meet two fundamental challenges that are part and parcel of 5G. As the network will run at increasingly higher frequencies than the current 4G bands, signals will be much less able to ‘bend’ around obstacles such as hills, buildings or even bus shelters. Consequently, technology will rely on direct line of sight between radio antennae.

A highly accurate 3D model of the urban landscape is therefore needed to optimally place 5G assets. Secondly, weather will have an impact on network performance, as raindrops can degrade signal strength, so it is critical to understand the degree to which the day-to-day variability of UK weather, including extremes, must be designed into the future networks.

To understand more about these challenges, a team of experts, led by the Ordnance Survey, was brought together to create detailed 3D modelling of a cityscape. The team included the Met Office and the 5G Innovation Centre (5GIC) based at the University of Surrey in Guildford.

Getting the full picture

The project concentrated on Bournemouth, with Ordnance Survey building a ‘digital twin’ of the area to a 10 cm granularity. The Met Office team was led by Dave Jones, Head of Observations Research and Development.

“The project just happened to be looking at the frequency band that I studied for my PhD, so it was very exciting to work on it,” explains Dave.

The Met Office already had a lot of the science in place to model rainfall distributions and understand their impact on 5G frequencies. To make the model as realistic as possible, Dave’s team used real weather scenarios from the Met Office weather radar archive, focusing on high profile events such as Storm Angus from November 2016.

Archive data was available at a spatial granularity of 2 km, but the team used a technique known as downscaling to represent how rainfall varies on a scale of 100 m, which is representative of the sorts of antenna separations we might expect for 5G.

“The project just happened to be looking at the frequency band that I studied for my PhD, so it was very exciting to work on it.”

The work highlighted two likely worst-case scenarios for 5G: summer convective downpours and frontal conditions with temperatures just above freezing, where rainfall might arrive at ground in the form of melting snow. Both instances could severely affect signal strength, depending on the frequency bands that are eventually selected.

Knowing the details of how weather affects the signals, providers will need to weigh up cost versus availability.

“The rainfall events will reduce the capacity of the network, but do we need a network to be working at full capacity when there’s a downpour? Will people be streaming video outside in the pouring rain?” explains Dave.

The work is extremely important for UK businesses as 5G could well transform people’s day-to-day lives. From augmented reality to autonomous cars, lots of technology will rely on the connectivity that 5G will unleash. The Met Office’s involvement in the project will contribute to optimum investment in the networks.

The project is an excellent example of the Met Office working in close collaboration with other leading organisations, including the Ordnance Survey and 5GIC. “Working together with subject experts from different fields was really rewarding,” says Dave. “It was good to work with such a focused team.”

From signal to forecast

5G communications is going to be part of the crucial infrastructure for the UK and the Met Office is keen to play a part in that. However, there’s one other possibility that the link between 5G and rainfall opens up. Because we know that rainfall affects signal strength, then we can turn that around and use real-time 5G signal strength to tell us about rainfall rates as it is falling.

“People have already been trying this with 4G, particularly in the Netherlands,” explains Dave Jones, Met Office Head of Observations Research and Development. The technology will never replace the Met Office’s rainfall radar network, but the readings could complement the Met Office’s existing observations networks to give an even greater level of detail on surface rainfall in urban areas.

This is just one several possible opportunistic observations that the Met Office is exploring to improve our understanding of the state of the atmosphere from infrastructure that was originally designed for other purposes.