Granted Patent B2

US 12,739,682 · App. 18/413,793 · Granted Sep 15, 2026

Inventors: Umur Karabulut (Munich, DE); Ahmad Masri (Tampere, FI); Anupam Shrivastava (Dallas, TX); Mustafa Cemil Coskun (Murray Hill, NJ)

Assignee: NOKIA TECHNOLOGIES OY

H04W24/10H04W36/0083

Abstract

Claims (23)

1 . An apparatus comprising: at least one processor and at least one memory storing instructions that, when executed by the at least one processor, cause the apparatus at least to:

receive an indication of a first beam index and a second beam index;

determine, at a user equipment, a difference between signal strength for a first beam of a serving cell of a network and signal strength for a second beam of the serving cell,

wherein the first beam of the serving cell comprises a channel state information reference signal and the second beam of the serving cell comprises a synchronization signal or a physical broadcast channel,

wherein the first beam and the second beam are determined based on the indication of the first beam index and the second beam index;

determine a timing advance value of the serving cell and a timing advance value of at least one non-serving cell;

provide the determined timing advance values and the determined difference between the signal strength for the first beam of the serving cell and the signal strength for the second beam of the serving cell as inputs for a machine learning model, wherein an output of the machine learning model is numerical data or categorical data, wherein the numerical data comprises a probability value, wherein the machine learning model is used for measurement reporting;

determine, based on the output of the machine learning model, that a measurement report should be provided to the network,

wherein the measurement report comprises an indication of the output of the machine learning model and an indication of the determined difference between the signal strength for the first beam of the serving cell and the signal strength for the second beam of the serving cell,

wherein determining that the measurement report should be provided to the network comprises comparing the output of the machine learning model to a threshold value;

provide the measurement report to the network;

receive a configuration from the network to train the machine learning model at the user equipment;

use the determined timing advance values and the determined difference between the signal strength for the first beam of the serving cell and the signal strength for the second beam of the serving cell to train the machine learning model at the user equipment;

determine, at the user equipment, a difference between signal strength for a first beam of at least one non-serving cell and signal strength for a second beam of the at least one non-serving cell;

determine a timing advance value of the serving cell of the network and a timing advance of the at least one non-serving cell;

provide the determined difference for the serving cell and the determined difference for the non-serving cell and the determined timing advance values as inputs for a further machine learning model,

wherein the output of the further machine learning model is numerical data or categorical data,

wherein the further machine learning model is configured for triggering conditional handover execution and is independent from the machine learning model;

determine, based on the output of the further machine learning model, that a conditional handover procedure should be performed,

wherein determining that the conditional handover procedure should be performed comprises comparing the output of the further machine learning model to a threshold value;

perform the conditional handover procedure;

receive a configuration from the network to train the further machine learning model at the user equipment; and

use the determined difference and the timing advances values to train the further machine learning model at the user equipment.

Assignments (6)

Priority Claims (1)

Continuity (1)

References Cited (32)