putilov_denis - stock.adobe.com
How AI can improve telecom RAN operations and analytics
Telecom operators can harness AI to improve their RAN operational efficiency and analytics, which enables them to improve customer experience and network management.
Telecom operators are tightening their embrace of AI to transform their analytics capabilities and streamline radio access network operations.
The advent of generative AI has irrevocably altered the telecom industry. Telcos and their ecosystem partners are racing to use large language model-based tools to improve overall experiences in areas such as call center customer support, customer engagement and bill explanation.
At the same time, generative AI is leading telcos to shift from long-established AI and machine learning technologies. ML, in concert with AI, continues to play a key role in telco decision-making. But, as traditional analytics methods struggle to cope with the vast amount of data telecom operators obtain from diverse sources, this new generation of AI-driven predictive analytics provides telcos with the ability to extract actionable insights.
AI-driven analytics: Improving telco analytics outcomes
AI-driven analytics can enable telecom operators to do the following:
- Gain a better understanding of customer behavior.
- Predict network congestion.
- Anticipate service demand fluctuations.
With these advanced algorithms, telecom operators can optimize network resource allocation, enhance service quality and address network issues before they affect end users and customer satisfaction. They can also better manage churn. By employing advanced analytics, for example, telcos can prioritize how to retain customers who have reported unsatisfactory network experiences.
Additionally, AI algorithms give operators the ability to comb through historical data more efficiently to predict network failures before they occur. This includes embedding AI and ML models within digital twins that can detect anomalies with incoming telemetry and trigger alerts to address emerging issues. AI-infused digital twins can also help telcos allocate resources more efficiently, pinpointing parts of the network that need infrastructure upgrades and modifications.
These AI-enabled capabilities have a direct effect on emerging AI radio access network (RAN) strategies.
AI RAN: Paving the way for telco RAN innovation
The increased interest in AI RANs is fueled heavily by the integration of AI capabilities in operator RAN portfolios. This integration enables mobile network operators to improve RAN efficiencies throughout development and deployment. It also benefits the overall mobile network and augments network edge intelligence.
In addition, AI is being tapped to usher in a new era of RAN performance, optimizing transmission power and antenna configuration, among other factors. Digital twins, meanwhile, are playing a critical role in AI RAN product development, bolstering support of use cases such as remote control, monitoring and analysis, and scenario planning.
Other AI use cases for the RAN include the following:
- Finding ways to cut energy consumption through better power allocation management.
- Finding more energy-efficient configuration settings for network modes.
- Pinpointing energy-hungry components.
Harnessing AI RANs can boost telco AI as operators advance their digital transformation through greater mobility, flexibility, security and reliability. It also helps telcos strengthen their sustainability missions by making existing implementations more intelligent.
AI RANs will gain additional momentum from industry groups, such as the AI-RAN Alliance, which debuted at this year's Mobile World Congress. AWS, Arm, DeepSig, Ericsson, Microsoft, Nokia, Northeastern University, Nvidia, Samsung Electronics, SoftBank and T-Mobile are among the alliance's founding members.
Ron Westfall is research director at The Futurum Group, covering 5G, cloud computing, security and more. He has covered the digital and IT markets for over 20 years.
