AI-Powered Network Monitoring for 5G Stand-Alone Infrastructure let Networks achieve Extended Efficiency

Contributed by Claudio Romani, CEO, Resi S.p.A

With the increasing complexity of telecommunications networks and the progressive virtualization of crucial network functions, network monitoring stands out more than ever as a critical IT process for any company, especially those in the telecommunications sector characterized by highly complex distributed architectures.

To an operator, suboptimal responses to service unavailability, delays in activating new customers and, more generally, the inability to promptly understand the evolution of customer’s needs and habits are risk factors leading to loss of competitiveness and direct business implications. Therefore, proactive service assurance in telcos must be capable of anticipating potentials issues, criticalities and trends before their impacts occur.

Meet Claudio, and the Resi S.p.A team, at FutureNet MENA 2024

In order to implement the holistic approach needed to create a unitary vision from fragmented views of underlying architectures, monitoring systems are gradually evolving thanks to the use of artificial intelligence (AI) and machine learning, so to enhance and complement existing network monitoring systems based on deterministic and statistical approaches. In this context, Stand-Alone 5G (SA5G) networks represent the privileged field of research and innovation for new functions.

In addition to processing metrics of end-user service quality (Quality of Experience, QoE), evolved network traffic monitoring and analysis systems for SA5G networks leverage AI-based expert systems implementation and machine learning techniques to deliver two major functions: event Detection and Prediction of future traffic trends.

Detection functionalities enable the identification of single events or anomalous trends to highlight or anticipate issues in proper functioning, performance, and system security. These functionalities correlate metrics from different network nodes and segments (access, core, etc.) and different types of traffic (signaling, data, voice, etc.).

Prediction functionalities integrate traffic analysis by identifying and isolating trends from transient fluctuations (noise), predicting their impacts on the infrastructure in order to promptly adapt resources to future service needs (network planning, smart capex, cloud scaling, etc.).

Detection and Prediction modules will work synergically to feed external automated network supervision and provisioning (Network Operation Automation) systems through standard interfaces (APIs, Kafka, etc.).

AI-driven Quality of Experience monitoring can feed decision-making processes and supporting tools (Business Intelligence, Marketing, etc.) regarding customer network usage, focusing on the impacts that emergence of new services can have on users. Timely interpretation of trends and behaviors allows for adjusting operational processes, effectively planning the launch of new services (time-to-market), and improving customer satisfaction by fostering customer loyalty.

A Further step in the evolution of service assurance system would be the automated acquisition of information from external systems capable of identifying planned events (sports, cultural or social events, holidays, etc.), predictable events (weather conditions), or unforeseen events (disasters, sudden fluctuations in air traffic, inaccessibility of certain services or infrastructure) capable of altering traffic usage, volumes, and geographical location. Such knowledge would provide the tools to react to external stresses affecting the services, Including threats to strategic infrastructures reported by competent authorities.

By exploiting AI techniques, evolved monitoring systems will harness the vast amount of data from the mobile network, integrating it with public data from external sources and expertise gained in human skills transferred to automated learning systems. This will enable the implementation of the concept of “extended network efficiency,” integrating infrastructure and service optimization with the pursuit of financial (cost-saving, steering), socio-cultural (social responsibility), and strategic (infrastructure security) objectives.