What is the relationship between AI and 5G?

The impact of 5G

Commercial deployment of 5G is underway. But put simply, 5G isn’t just another G. It’s a whole ecosystem shift in how networks are run and managed, including how apps run on the network. .

There are three main groups of use cases in 5G:

  • Enhanced Mobile Broadband, or EMBB
  • Fixed wireless access that uses the millimeter wave spectrum
  • Fiber, where the most common use is cable for our home broadband.

Other groups of emerging use cases include massive machine-type communication, or MTC.

This is where the connectivity and density of 5G really comes into play.

MTC enables connectivity for large numbers of devices – millions, billions of devices in fact, all of which are connected. While they’re more likely to send very low data rates, the sheer number of devices and long battery life means they can open the doors to whole new industrial use cases. For example, surveillance, agriculture, agriculture, transportation, automotive, smart cities, and healthcare could all transform with MTC. It’s about connecting human expertise to a large number of connected sensors for faster and more efficient information.

Another emerging technology is ultra-reliable low-latency communications, or URLLC. This is where 5G shines. Use cases with URLLC can provide very low latencies, down to one millisecond, which is a perfect solution for critical use cases – vehicle-to-vehicle, remote diagnostics or remote surgery.

URLLC’s low latency is perfect for critical use cases such as those in healthcare.

How is AI involved in 5G?

When it comes to 5G networks, AI is no longer an asset, but an indispensable component to deal with the enormous complexity that comes with 5G. AI, and the data and automation capabilities that come with it, support the diverse ecosystem of evolving networks in ways that humans alone cannot.

Expectations of 5G are high due to its potential to transform industries. Service providers expect the high performance, low latency, throughput and availability promised by 5G. As a result, the ability to exploit 5G networks will need to accelerate – in fact, the development of high-level operational capabilities such as contactless and self-healing networks is already underway to meet this growing demand.

Main challenges

The evolution of networks involves difficult challenges, the first of which relates to data, specifically, how to shape network operations to be data-centric and data-driven. For example, data elements within a 5G network are highly distributed. It comes in all shapes, sizes and volumes. So how is it possible to effectively manage this data? After all, it’s data that drives features like machine learning and advanced analytics. Without it, we cannot operate future networks.

First, a clearly defined and executed data-driven strategy is crucial for service providers; the one who determines how data should be managed end-to-end, from ingestion to final decision-making.

Second, clear decisions need to be made about where and how data is processed, so that AI logic can make timely decisions. For example, data could be transferred to a centralized cloud location to be processed for AI inference, but this can lead to high transfer costs and additional delays, especially for real-time use cases. where decisions have to be made in a fraction of a second. Instead, AI inference could be brought closer to the data source and create a shorter, leaner pipeline.

Another important aspect is ensuring data quality and lineage, end-to-end, so that decisions can be made based on reliable, high-quality data. It makes no sense to rely on AI logic if the data is corrupt.

And finally, organizational transitions around skills, technology development, and the sustainability of employee skills are all additional challenges that can arise with the adoption of 5G and AI.

Listen to the podcast

How to address the challenges of 5G and AI adoption

To overcome these new challenges, Ericsson has changed its approach – from being reactive to becoming much more proactive and predictive, which is the basis of our AI modeling. The result is a model called Ericsson Operations Engine. Along with our data-driven approach, we are also developing our people who can see the network from an end-to-end perspective.

We also focus on data analysis, skills development and 5G technologies, as well as the development of specific use case experts to meet new industry requirements. We need to have the skill to understand the whole ecosystem of these emerging use cases – not just an understanding of the technology, but also of the different platforms and tools to help manage network operations in a more fluid and more automated.

The Perspectives of AI and 5G: Network Slicing and Intent-Based Operations

As service providers begin to offer 5G services to enterprises, efficient network utilization will be crucial for them to reduce costs. This is where network slicing comes in.

Network slicing is a unique technology in 5G, where the network can be logically sliced ​​end-to-end to deliver customizable service performance. Service providers will be able to segment the network and offer different segments of the network to different enterprise customers and ensure that they receive the level of performance for which they pay.

AI-Driven 5G Network Slice Operations


Of course, slicing comes with its own challenges, including its technical complexity and the fact that it’s a cross-domain domain. Therefore, we have worked on several advanced AI techniques to help customers prepare for the challenge of operating such complex systems. The future of networks ultimately lies in cognitive systems, where networks can apply a combination of machine learning and machine reasoning, which is built from a knowledge base and reasoning engines to generate conclusions.

This technique is not only relevant for network slicing, but for all complex network operations, because we often don’t have enough data or enough labels to form all possible scenarios. This approach, however, allows the machine to learn on its own and make critical decisions on its own without first being trained or instructed to do so.

We believe that this whole approach, along with intent-based operations, will be a crucial step in making 5G operations as autonomous as possible. Exciting times are upon us.

Want to know more?

This is how network operations can make 5G systems resilient.

Learn all about monetizing 5G with intent-based network operations.

Learn more about managed services.

Comments are closed.