AI, SON and the self-driving cellular network

Artificial Intelligence (AI) is being groomed, says Rethink Technology Research to partner with self-optimising networks (SON) to create cellular networks that know the user, perform brilliantly and defy complexity. But 50% of network providers worry they will not be able to attract sufficient AI skills.

While everyone today is aware of the race to build a self-driving car, the race to build a self-driving cellular network is far less of a public property, but nonetheless, it is a race that is critical to the survival of existing mobile network operators (MNOs).

The rise in self-optimising networks, along with a huge increase in the total number of cells, will lead to a radical uptake of virtualised Software Defined Networks. This will make it possible to dial network resources up and down on-demand, but only if MNOs can cope with another level of network complexity.

Rethink has once again consulted all of its MNO contacts in MNO strategy departments all over the world, and found how they plan to deal with this complexity. Essentially it is to move towards automated network optimisation or (SON) and at the development of AI and machine learning tools in parallel, to guide SON efforts, end to end, across their networks.

As soon as 5G mobile networks begin in anger, they will be extremely automated, and this step will mean huge shipments of automated SON tools which will take on the critical role it has always previously fallen short of. The entire thing must be far more intelligent, Rethink maintains, and the introduction of AI/ML will help operators automate their networks in a highly responsive way, to create what we are calling ‘the self-driving telco.’

Artificial Intelligence and machine learning (ML) for networks is being accelerated through standards bodies as we speak. ETSI will develop use cases and requirements for telco network AI, which it will model in 2018 and then move on to testing and automation in 2019-2020, to be ready just in time for the first 5G self-driving networks. AI will also be developed to detect cyber-crime and suspicious behavior in parallel.

While we can see this is not an overnight process, because the technologies and processes are immature, the operators who start early will get the greatest benefit. For many MNOs, this will be part of a wider progress towards SDN and 5G, which will take place over as many as 10 years.

AI and machine learning are, of their nature, slow to perfect – because the models need to learn until they reach the stage where they can make quicker, better decisions than a team of trained experts. Once automated decision making is in place it will offer the side benefit of helping an operator to understand its customers in greater depth, and the same goes for network planning and traffic management.

These are also areas where expert employees are in short supply, and there are huge volumes of information to tap into. AI can be used to ensure customers are receiving consistent quality of service by allocating resources where needed, and fixing problems. It can also be used to identify malware, hacks and suspicious user behavior.

To find more about AI and the Self Driving Cellular Network and to see our Forecast of the technology players which it will benefit, buy a copy of our Ran Research paper “The Self Driving Cellular Network.” The report shows how those operators who can attract AI expertise will lead the shift into truly intelligent networks.

Our survey showed the main perceived barriers to integrating AI/ML into a SON platform to improve intelligent automation. The first is getting enough of the right kind of people, with the right skills. This was placed among the top two problems by 51% of respondents. This is followed by the related factors of the cost of building the AI engine, and the uncertain Return On Investment. These factors will drive, as with SON overall, an increasing reliance on hosted services, and AI/ML engines in the cloud will be important ways for many operators to accelerate adoption and reduce risk.

When it comes to end-to-end automation, including core, backhaul, RAN and other elements, such as edge compute servers, and even some devices, there is even greater caution. Only 14% expect to have automated 60% of planning, optimisation and MANO processes by the end of 2020, and in 2025, 39% will still have automated fewer than 40%.

Methodology

The wireless forecast included in this report is based on research on the top 40 international mobile operator groups, which account for 80% of the global mobile subscribers (IMG-40). From this representative group of operators, the macrocell and metrocell forecasts are developed.

From the starting point of a calculation of the number of cell sites already deployed worldwide, forecasts were made of the numbers of base stations that would be rolled out to brand new sites and to replace or upgrade existing sites. These deployment forecasts were then categorised by network topology, spectrum band, spectrum mode, region and other factors. The equipment deployed in each case was also surveyed and modeled.

The term “Self-Driving Networks” has been claimed as a trade mark by Juniper Networks, and used in its marketing. Our report is about the generic concept referred to by multiple vendors as a hybrid between SON and AI.

Comment on this article below or via Twitter: @ VanillaPlus OR @jcvplus

RECENT ARTICLES

Phoenix Tower International gains investment from Grain and BlackRock

Posted on: March 29, 2024

Phoenix Tower International (PTI) has announced that Grain Management (Grain), through its flagship funds, and BlackRock, through a fund managed by its Diversified Infrastructure business (BlackRock) have made an investment

Read more

Connectbase expands baltic connectivity with Bitė partnership

Posted on: March 28, 2024

Connectbase has announced the addition of Bitė to its ecosystem. This partnership marks a step forward in enhancing connectivity options within the Baltic region, providing a link between local and

Read more