Networks of the future will be fuelled by analytics, AI, and 5G

Artificial Intelligence (AI) and 5G are red hot topics today. However, despite all the hype and discussion around connected and intelligent applications, there is still a lot to be said about how these two technologies will leverage each other to deliver the networks and associated use cases of the future. By Brian Lavallée, Submarine Networking Solutions Expert, at Ciena.

  • 4 years ago Posted in

When compared to existing 4G LTE networks, 5G will offer unprecedented speeds, much lower latency, higher connectivity, and higher availability to power smart cities, connected and autonomous vehicles, AR/VR streaming, and numerous other advanced applications.

AI, fuelled by analytics, has the potential to help improve the efficiency of network communications and maintenance, while safeguarding network uptime through automated policy-based decision making through streaming real-time network data. It also has the potential to identify and prevent potential service disruptions, detect suspicious network behaviour for increased security, and proactively improve overall network reliability. This combination represents significant value for operators preparing to adopt emerging technologies, such as 5G, as they mature.

AI and 5G have the potential to help solve the challenges that impede network evolution. Together they could reduce costs related to maintenance and network downtime and improve decision-making on bandwidth allocation and network repairs to ensure high-availability and performance for users across the globe.

AI: Challenges vs. prospects

For network operators, understanding the value of AI must start with an understanding of AI itself and what it can offer in terms of network management, efficiency, profitability, and security.

AI is a “thinking” machine capable of monitoring network behaviours, identifying potential problem areas, detecting what is “not normal”, and taking corrective action. Harnessing the power of data – analytics, AI can introduce greater efficiency to the workflow by removing the requirement for human operators to make basic decisions on migration and path selection. This will optimise network functionality and enable bandwidth on demand.

Despite the apparent benefits of ‘automated networks’, it’s clear that operators will not surrender complete control of their networks– instead opting for adaptive networking practices that harness the power and efficiency of data-driven AI and combine it with the invaluable experience of their engineers.

AI should be implemented in controlled stages, with rigorous testing to ensure it has both enough data and the right data to form solid decision-making policies. By developing projects individually, operators can ensure they are satisfied with the results in one area, before scaling up deployments to cover more network functions, thereby reducing the potential for errors ahead of broader adoption.

AI’s role in 5G

It is predicted that there will be one billion users of 5G by 2020 with one in seven mobile connections made via 5G by 2025. And, while 5G is still in the testing phase, the initial rollouts seen in South Korea, the US, and the UK, will be used as proof points for how the technology will enable operators to determine the most cost-effective models upon which to expand to national and regional coverage.

AI won’t just be beneficial to 5G, it will be a necessity. By drawing on the massive influx of data to form superior policies, AI can be used to examine network activity and suggest the appropriate action in the event of service disruptions. In addition, AI will enable self-healing capabilities. Through real-time data analysis, AI will compress decision-making timelines by orders of magnitude, repairing or even reconstructing the network in a matter of minutes to minimise disruptions from damaged cables or attempted network intrusions. The potential savings through the prevention of revenue loss will be a crucial factor in ensuring cost-effective 5G services as operators evolve their networks over the next decade.

It is also likely that the increase in device traffic on 5G networks will usher in a greater threat to operators from Distributed Denial-of-Service (DDoS) attacks, as hackers will be able to launch attacks of unprecedented size. AI will offer increased security through proactive network monitoring, using historical data to spot anomalies on network services and signs of intruder connections, and taking action to protect the network and preserve functionality.

Therefore, it makes sense for operators to start integrating AI capabilities into their existing 4G network infrastructure now, where it makes sense, to provide the additional functionality that will be a must for managing independent 5G networks in future.

AI at 5G speed

There is no doubt AI and 5G will soon take on a significant role in mission critical telecommunications services. As operators define the most effective models and connect individual deployments to create broader networks, AI-enabled 5G will evolve to uncover exciting new possibilities and use cases. Together these technologies will drive greater business profitability, innovation and enhanced user experiences. 

For operators, the time to lay the groundwork for 5G connectivity is now. A big part of that journey will involve developing AI capabilities that go beyond today’s network functionality and embrace the world of high-bandwidth, always-on connectivity, to give rise to intelligent infrastructure and widespread connected device networks across the globe.

820 words

 

By Shirin Esfandiari, Product Director at Oracle.
By Laurent Bouchoucha, VP Business Development Network Solutions - Alcatel-Lucent Enterprise.
Nicolas Roussel, Technical Manager at Siemon, discusses whether to select singlemode or multimode...
If you’re wondering why digital transformation still isn’t happening and you’ve interrogated...
The COVID-19 pandemic has helped unify the call for universal broadband to ensure that no family,...
By Michael Davies, Director of WAN Services Product Management at GTT.