AI modeling for remote electrical tilt: a live discussion

Mobile operators continuously make remote electrical tilt (RET) adjustments to optimize mobile network performance, maximize coverage and capacity, and provide great customer experiences.

Rakuten Mobile, Inc. Chief Advisor to the Chief Data Office, Dr. Petrit Nahi joined us on Zero-Touch Telecom Live this week to discuss building an RET AI model, his experience running it and the improvements that have been made over time to improve network performance. Check out the replay below.

RET is traditionally conducted on a weekly basis, often with incomplete data, to make best-effort network performance improvements based on factors like traffic, subscriber movement and signal strength in a particular area.

In our interview, Petrit detailed how AI helped transform Rakuten Mobile's approach to RET. This includes collecting real-time data in millisecond intervals across all network nodes and using it to rapidly adapt to network changes on a daily basis.

Also covered during the discussion:

• The evolution of network optimization from GSM to UMTS and how AI took on the role of managing antennas.
• Data collection challenges that arise across multiple vendors and steps to overcome them.
• The significant improvements achieved by Rakuten Mobile, including a 25% increase in spectral efficiency and improved handover success rates from 97% to 99%.

If you haven’t already, be sure to read our Zero-Touch Telecom newsletter article on this subject, where we dive deeper into our experience.