US 2025/0167905 A1 · TECHNIQUES FOR BEAM CHARACTERISTIC PREDICTION USING FEDERATED LEARNING PROCESSES

Abstract

Methods, systems, and devices for wireless communications are described. A first network node, such as a user equipment (UE), may receive a first model configured to predict at least one of time-domain characteristics or spatial-domain characteristics associated with a set of channel measurement resources (CMRs). The first network node may generate first measurement information corresponding to a first quantity of CMRs of the set of CMRs and may input the first measurement information into the first model. The first network node may obtain, as an output of the first model, first predicted information corresponding to a second quantity of CMRs of the set of CMRs, receive signals using the second quantity of CMRs, and generate second measurement information corresponding to the second quantity of CMRs. The first network node may train the first model with the second measurement information and transmit the trained first model to a second network node.

Claims30

Dependent claims

A first network node for wireless communication, comprising:

a memory;
and
at least one processor coupled to the memory, wherein the at least one processor is configured to:
receive a first model configured to predict at least one of time-domain characteristics or spatial-domain characteristics associated with a set of channel measurement resources;
generate first measurement information corresponding to a first quantity of channel measurement resources of the set of channel measurement resources, wherein the first quantity of channel measurement resources correspond to a first set of signals;
input the first measurement information into the first model;
obtain, as an output of the first model, first predicted information corresponding to a second quantity of channel measurement resources of the set of channel measurement resources, wherein the first predicted information includes at least one of:
one or more predicted time-domain parameters or one or more predicted spatial-domain parameters;
and
receive a second set of signals, wherein the second quantity of channel measurement resources correspond to the second set of signals.

7 more elements

A first network node for wireless communication, comprising:

a memory;
and
at least one processor coupled to the memory, wherein the at least one processor is configured to:
transmit signals within a set of channel measurement resources;
receive, from a second network node, a first trained model associated with at least a first portion of the set of channel measurement resources, the first trained model configured to predict at least one of time-domain characteristics or spatial-domain characteristics associated with the set of channel measurement resources;
receive, from a third network node, a second trained model associated with at least a second portion of the set of channel measurement resources, the second trained model configured to predict at least one of time-domain characteristics or spatial-domain characteristics of the set of channel measurement resources;
generate a third model based on the first trained model and the second trained model, the third model configured to predict at least one of time-domain characteristics or spatial-domain characteristics associated with the set of channel measurement resources;
and
transmit, to at least one of the second network node or the third network node, an indication of the third model.

6 more elements

Dependencies are inferred from the claim text.

Techniques for beam characteristic prediction using federated learning processes

Life of this patent

Techniques for beam characteristic prediction using federated learning processes

Life of this patent

Abstract

Claims30