Nokia has added new machine learning features to its Analytics Services portfolio, allowing operators to improve network management using real-time data.
The offering will now include the Mobility Analysis and Optimisation capability, which captures and analyses routine measurements such as signal strength from subscriber devices to reduce dropped calls.
The new Spectral Performance Management feature uses the same feed of end-user data to plan capacity needs of the network.
Nokia has also added the Cell Site Degradation Prediction service, which collects metrics such as temperature at cellular sites that have experienced failures and uses these to preempt future problems in others.
In addition, the VoLTE Audio Gap Analysis service applies machine learning to 4G calls to identify the root cause behind voice problems.
Finally, the new Predictive Video Analytics service will track patterns in encrypted traffic on apps such as YouTube and Netflix to look for bottlenecks that are damaging video quality.
According to Nokia, the new offerings can boost first-time resolution of network issues by 20 to 40 percent and reduce dropped calls by up to 35 percent.
Built on Nokia’s AVA knowledge library, the Analytics Services portfolio is offered through the cloud to eliminate capex requirements, and uses a DevOps model to add quick updates.
Dennis Lorenzin, Head of Network Planning and Optimisation, Global Services, Nokia, said: “Our analytics services help to cope with the complexity of today’s networks.
“We can augment human intelligence to improve efficiency and reduce the cost of operations. In addition, we can provide deeper insights to improve quality of experience based on subscriber, device and application usage patterns.”
Sheryl Kingstone, Research Director at 451 Research, added: “The business potential of analytics is yet to be tapped fully by the telco industry. Telecom operators often require help to unlock the value of data, through targeted and actionable insights.”