Method and system for beam alignment in wireless network

Patent metadata

Field	Value
Publication number	US-12425875-B2
Application number	US-202217956318-A
Country	US
Kind code	B2
Filing date	Sep 29, 2022
Priority date	Mar 29, 2020
Publication date	Sep 23, 2025
Grant date	Sep 23, 2025

Abstract

The disclosure provides methods and apparatuses for performing beam alignment. In some embodiments, a method of performing beam alignment by a user equipment (UE) of a wireless network, includes determining, by the UE based on a plurality of parameters associated with the UE, a first quality level of a connection of the UE in a first orientation and at a geographical location. The connection uses a first beam from at least one network entity of the wireless network. The method further includes determining, by the UE based on the first quality level, a second orientation from a plurality of orientations at the geographical location. The second quality level of the second orientation exceeds the first quality level. The method further includes changing, by the UE, an orientation of the UE from the first orientation to the second orientation.

First claim

What is claimed is:

1. A method of performing beam alignment by a user equipment (UE) of a wireless network, comprising:
determining, by the UE based on a plurality of parameters associated with the UE, a first quality level of a connection of the UE in a first orientation and at a first geographical location, the connection using a first beam from at least one network entity of the wireless network;
determining, by the UE based on the first quality level, a second geographical location that maximizes a data rate of the UE and is within a predetermined distance threshold from the first geographical location;
determining, by the UE based on the first quality level, a second orientation from a plurality of orientations at the second geographical location, wherein a second quality level of the second orientation exceeds the first quality level; and
changing, by the UE, an orientation of the UE from the first orientation to the second orientation, and causing a geographical location of the UE to change from the first geographical location to the second geographical location,
wherein the determining of the second orientation comprises:
estimating, by the UE based on the plurality of parameters, for each orientation of the plurality of orientations, a corresponding quality level of the connection at the first geographical location;
identifying, by the UE and using a machine learning (ML) model, the second orientation based on the corresponding quality level of the connection; and
determining, by the UE, that the corresponding quality level of the second orientation exceeds the first quality level, and

wherein the ML model has been trained for identifying the second orientation by:
receiving one or more parameters corresponding to a particular beam;
normalizing, based on the one or more parameters, a reference signal received power (RSRP) of the particular beam;
determining, based on the normalized RSRP, an upper confidence bound (UCB) value for a candidate geographical location corresponding to the at least one network entity and the first orientation of the UE;
storing the UCB value in a database of the UE; and
updating, based on stored UCB values, the candidate geographical location having the at least one network entity with a beam and the second orientation.

2. The method of claim 1 , further comprising:
maintaining, by the UE, the connection with the at least one network entity while the orientation of the UE is changed from the first orientation to the second orientation; and
performing, by the UE, communication with the at least one network entity in the second orientation at the second geographical location.

3. The method of claim 1 , wherein the plurality of parameters comprises at least one of a signal-to-noise ratio (SNR) of a beam received from the wireless network, a signal-to-interference-plus-noise ratio (SINR) of the beam, a signal strength, a radio link failure frequency, a width of the beam, a beam identifier (ID), a sector ID, a public land mobile network (PLMN) ID, a physical cell ID, a RSRP of the beam, a reference signal received quality (RSRQ) of the beam, a longitude of the UE, a latitude of the UE, an azimuthal angle of the UE, a polar angle of the UE, and a height of the UE from a ground.

4. The method of claim 1 , wherein the UE is connected to the at least one network entity using the first beam or another beam.

5. The method of claim 1 , wherein the estimating of the corresponding quality level of the connection comprises:
monitoring, by the UE based on the plurality of parameters, the corresponding quality level of the connection for each orientation of the plurality of orientations over a period of time;
identifying, by the UE, geographical locations associated with the plurality of orientations of the UE; and
storing, by the UE in the database of the UE, information about the corresponding quality level of the connection and the geographical locations associated with the plurality of orientations.

6. The method of claim 5 , wherein the information about the corresponding quality level of the connection comprises an UCB value for each of the geographical locations associated with the plurality of orientations of the UE.

7. The method of claim 1 , wherein the identifying of the second orientation comprises:
identifying, by the UE, the first geographical location of the UE that establishes the connection with the at least one network entity in the first orientation;
determining, by the UE based on information about the first quality level of the connection stored in the database of the UE, the candidate geographical location having one or more network entities with a beam nearest to the geographical location of the UE; and
determining, by the UE, the second orientation based on the candidate geographical location and the information about the first quality level of the connection.

8. A method of performing beam alignment by a cloud server of a wireless network, comprising:
receiving, by the cloud server from a user equipment (UE) of the wireless network, a plurality of parameters associated with the UE;
determining, by the cloud server based on the plurality of parameters, a first quality level of a connection of the UE in a first orientation and at a first geographical location, the connection using a first beam from at least one network entity of the wireless network;
determining, by the UE based on the first quality level, a second geographical location that maximizes a data rate of the UE and is within a predetermined distance threshold from the first geographical location;
determining, by the cloud server based on the first quality level, a second orientation from a plurality of orientations at the second geographical location, wherein a second quality level of the second orientation exceeds the first quality level; and
configuring, by the cloud server, the UE to change an orientation of the UE from the first orientation to the second orientation, and cause a geographical location of the UE to change from the first geographical location to the second geographical location,
wherein the determining of the second orientation comprises:
estimating, by the cloud server based on the plurality of parameters, for each orientation of the plurality of orientations, a corresponding quality level of the connection at the first geographical location;
identifying, by the cloud server and using a machine learning (ML) model, the second orientation based on the corresponding quality level of the connection; and
determining, by the cloud server, that the corresponding quality level of the second orientation exceeds the first quality level, and

wherein the ML model has been trained for identifying the second orientation by:
receiving one or more parameters corresponding to a particular beam;
normalizing, based on the one or more parameters, a reference signal received power (RSRP) of the particular beam;
determining, based on the normalized RSRP, an upper confidence bound (UCB) value for a candidate geographical location corresponding to the at least one network entity and the first orientation of the UE;
storing the UCB value in a database of the UE; and
updating, based on stored UCB values, the candidate geographical location having the at least one network entity with a beam and the second orientation.

9. The method of claim 8 , wherein the plurality of parameters comprises at least one of a signal-to-interference-plus-noise ratio (SINR) of a beam received from the wireless network, a signal strength, a radio link failure frequency, a width of the beam, a beam identifier (ID), a sector ID, a public land mobile network (PLMN) ID, a physical cell ID, a RSRP of the beam, a refere

Assignees

• Samsung Electronics Co Ltd

Inventors

• Agarwal Rishav
• Pooniya Ankur
• Roy Koustav
• Kadambar Sripada
• Paul Chirodeep

Classifications

• H04B17/328

  Reference signal received power [RSRP]; Reference signal received quality [RSRQ] · CPC title

• G06N3/092

  Reinforcement learning · CPC title

• G06N3/0464

  Convolutional networks [CNN, ConvNet] · CPC title

• G06N3/09

  Supervised learning · CPC title

• G06N3/08

  Learning methods · CPC title