Transfer Learning AI/ML Patent for Beam Management in Telecommunications

ChangeBridge: Patent Apps - AI & Computing (G06N)

Published April 2nd, 2026

Detected April 3rd, 2026

Summary

VIAVI Solutions Inc. filed a patent application (US20260094001A1) for transfer learning systems enabling AI/ML-based beam management in telecommunications networks. The invention allows pre-trained neural network models to be applied across different frequency bands to predict optimal beam configurations, reducing training requirements for 5G and future wireless systems. Application No. 18904464 was filed October 2, 2024.

View original document View source feed page

What changed

VIAVI Solutions Inc. received publication of a patent application covering transfer learning methods for AI/ML-based beam management in communication networks. The system generates neural network models designated with labels associating beam measurements from different frequency bands, trains the model using measurements from one frequency band, and applies the trained model to predict Top-K beam identifiers for another frequency band. This enables knowledge transfer between frequency ranges in telecommunications systems.

Patent applications do not create compliance obligations for other entities. Technology and telecommunications companies developing beam management systems may review the published application to understand VIAVI's claimed methods and assess potential licensing implications or design-around considerations. The application is informational and does not require external parties to take any action.

Source document (simplified)

← USPTO Patent Applications

TRANSFER LEARNING FOR ARTIFICIAL INTELLIGENCE (AI) AND MACHINE LEARNING (ML)-BASED BEAM MANAGEMENT

Application US20260094001A1 Kind: A1 Apr 02, 2026

Assignee

VIAVI SOLUTIONS INC.

Inventors

Yun CHEN, Onur Dizdar, Fehmi Emre Kadan, Stephen Wang

Abstract

Transfer learning (TL)-based systems, methods, and devices are provided for beam management in communication networks. In one aspect, a system may implement a transfer learning (TL)-based method comprising generating a neural network model for beam management in a telecommunications system, designating a plurality of labels, wherein one of the plurality of labels is associated with measurements from beams associated with a first set of measurements for a first frequency (f1). The system may also train the neural network model for the first frequency to produce a trained neural network model, including inputting measurements from beams associated with a second set of measurements for the first frequency (f1) and implement the trained neural network model to output a probability of each beam in the first set of measurements for the second frequency (f2) is a Top-1 beam, and determine beam identifiers (IDs) for the Top-K beams.