Pathloss prediction using a machine learning component

ChangeBridge: Patent Grants - AI & Computing (G06N)

Published April 7th, 2026

Detected April 8th, 2026

Summary

USPTO granted QUALCOMM Incorporated Patent No. US12598010B2 on April 7, 2026, covering a machine learning system for predicting pathloss values in wireless communication networks. The patent enables network nodes to determine signal attenuation across coverage areas using position information and map data processed through a machine learning component. The granted patent contains 20 claims protecting the method and system for ML-based pathloss prediction.

View original document View source feed page

What changed

USPTO issued Patent US12598010B2 to QUALCOMM Incorporated, granting exclusive rights to a method and system for predicting pathloss in wireless communication networks using machine learning. The patent covers techniques where a network node obtains position information for a subset of network node positions in connection with a map of a wireless coverage area, processes map information, and determines predicted pathloss values for additional network positions using a machine learning component.

For technology companies and telecommunications firms, this patent grant establishes QUALCOMM's intellectual property rights in ML-based pathloss prediction for wireless networks. Competitors developing similar machine learning approaches for optimizing wireless coverage, network planning, or signal prediction may need to evaluate potential licensing requirements or design-around strategies to avoid infringement on the 20 granted claims.

What to do next

Monitor for updates

Source document (simplified)

← USPTO Patent Grants

Pathloss prediction using a machine learning component

Grant US12598010B2 Kind: B2 Apr 07, 2026

Assignee

QUALCOMM Incorporated

Inventors

Thomas Markus Hehn, Tribhuvanesh Orekondy, Arash Behboodi, Ori Shental, Taesang Yoo, June Namgoong, Akash Sandeep Doshi, Ashwin Sampath, Juan Carlos Bucheli Garcia, Joseph Binamira Soriaga

Abstract

Various aspects of the present disclosure generally relate to wireless communication. In some aspects, a network node may obtain position information associated with a first subset of network node positions, of a set of network node positions, in connection with a map associated with a wireless coverage area. The network node may obtain map information associated with the map and may determine, based on a machine learning component, a first plurality of predicted pathloss values associated with a second subset of network node positions of the set of network node positions. Numerous other aspects are described.

CPC Classifications

G06N 3/02 G06N 3/042 G06N 3/044 G06N 3/0442 G06N 3/045 G06N 3/0455 G06N 3/0464 G06N 3/0499 G06N 3/08 G06N 3/088 G06N 3/09 G06N 3/091 G06N 3/092 G06N 20/00 G06N 20/10 H04B 17/347 H04B 17/3912 H04B 17/3913 H04L 25/0254 H04L 25/03165 H04L 2012/5686 H04L 2025/03464 H04L 41/145 H04L 41/16 H04L 45/08 H04W 16/22