Automation and AI: Key Enablers of Future Mobile Networks

Mobile network management had been a resource-driven, manual task starting from the introduction of first cellular services in 1970s until recent years. In early 2000s Mobile network operators (MNOs) designed, planned and optimized their networks with generic software tools, configured by hand, used for performance monitoring.

Now the telecom industry is at a turning point. Today’s networks are far too complex to manage manually and require use of advanced automation techniques that move beyond radio, spanning transport and core networks as well.

The complexities involved in 3G and early 4G deployments were lower compared to today’s deployments. When it comes to 5G, whether it is a standalone or non-standalone deployment, number of physical and virtual network resources will dramatically increase. Dynamic network policies will be needed for managing a continuously growing network with new added technologies such as small cells and beamforming. This dynamic infrastructure creates an environment that cannot be managed with conventional software tools. 

According to Gartner’s recently published hype cycle report (Hype Cycle for the Future of CSP Wireless Networks Infrastructure, 2018), 5G will be more complicated and a heterogeneous network (HetNet) will be commonly adopted, with a denser grid in hot spots, so topology changes must be planned. Operations need further automation and orchestration at scale as well. Achieving that goal requires a self-organizing network (SON) framework in place.

P.I. Works shares a similar vision with Gartner with regards to the benefits of network automation. Even more, P.I. Works’ view is that automating a highly complex, continuously growing infrastructure cannot be achieved without incorporating advanced AI algorithms.

Another prominent research company, ReTHINK, claims that AI combined with automation has the potential to slash network management costs with targeted resource allocation and planning. 

According to a recent survey conducted by ReTHINK across more than 100 MNOs, majority of the MNOs have a clear intention and plan in place to automate more than 59% of their radio network planning, management and optimization processes by 2025. Again, the same research indicates that 60% of the respondents are planning to automate more than 40% of their end-to-end processes by 2025, including radio, transport and core networks. 

As outlined in various analyst reports, there is a consensus across MNOs that automation powered by AI algorithms will play an important role in the evolution of mobile networks. 

P.I. Works, the provider of the most advanced and widely deployed network automation solution, is fully committed to driving the evolution of networks towards 5G. At Mobile World Congress 2018, Ooredoo Group announced the deployment of artificial intelligence solutions as part of a major partnership with P.I. Works to transform its network for 5G. There are many other operators that benefit from P.I. Works’ network automation solutions. 

P.I. Works’ Centralized SON, Smart Network Planning and Energy Saving products utilize known AI methods including but not limited to prediction and clustering. With those capabilities, P.I. Works product portfolio has already evolved into a more intelligent platform and has been outperforming the conventional automation solutions in terms of its impact on operational efficiency, network quality and customer experience. We are confident that the broader use of AI in network automation will yield greater benefits.