US 2026/0222246 A1Published application

Artificial intelligence machine learning models for channel sounding

AssigneeQUALCOMM Incorporated

InventorsYiyue CHEN, Sony AKKARAKARAN, Marwen ZORGUI, Sooryanarayanan GOPALAKRISHNAN

Life of this patent

Estimating the term…

Loading patent…

Abstract

Aspects of the disclosure are directed to artificial intelligence machine learning (AIML) models for channel sounding (e.g., Bluetooth channel sounding). In an aspect, phase measurement information associated with a channel sounding procedure is obtained. Root channel measurement information based on the phase measurement information is converted to AIML model input type(s), and provided as an input to AIML model(s), which output distance estimate information between an initiator device and a reflector device. In an aspect, the AIML model(s) include a first AIML model for feature extraction and a second AIML model configured as a regression model (e.g., instead of an end-to-end regression model). Such aspects may provide various technical advantages, such as improved ranging accuracy, particularly in multipath environments.

Claims20

Dependent claims

A device, comprising:

one or more memories;
one or more transceivers;
and
one or more processors communicatively coupled to the one or more memories and the one or more transceivers, the one or more processors, either alone or in combination, being configured to:
obtain phase measurement information associated with a wireless signal exchange procedure, the wireless signal exchange procedure comprising a first wireless signal transmitted by an initiator device to a reflector device and a second wireless signal transmitted by the reflector device back to the initiator device;
obtain root channel measurement information based on the phase measurement information;
convert the root channel measurement information to at least one artificial intelligence machine learning (AIML) model input type;
input to a first AIML model of a set of AIML models, the root channel measurement information;
and
obtain from a second AIML model of the set of AIML models and in response to the inputting of the root channel measurement information to the first AIML model, distance estimate information associated with a distance between the initiator device and the reflector device.

7 more elements

Dependencies are inferred from the claim text.