# US 12,113,590 B2 · Beam management method, apparatus, electronic device and computer readable storage medium

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 12,113,590 B2
- Document ID: US12113590B2
- Status: Granted
- Kind: B2
- Application number: 16/874,971
- Filed: 2020-05-15
- Published: 2020-11-19
- Granted: 2024-10-08
- Earliest claimed: 2019-05-16
- Assignees: Samsung Electronics Co., Ltd.
- Applicants: Samsung Electronics Co., Ltd.
- Inventors: Yiyan Zhang; He Wang; Songhui Shen
- Web page: https://patentgrep.com/patent/US12113590B2

## CPC

- H04B 7/0408: using two or more beams, i.e. beam diversity
- H04B 17/382: for resource allocation, admission control or handover
- G06N 3/044: Recurrent networks, e.g. Hopfield networks
- G06N 3/0442: characterised by memory or gating, e.g. long short-term memory [LSTM] or gated recurrent units [GRU]
- G06N 3/045: Combinations of networks
- G06N 3/08: Learning methods
- G06N 3/084: Backpropagation, e.g. using gradient descent
- G06N 3/092: Reinforcement learning
- G06N 20/00: Machine learning
- H04B 7/0617: {for beam forming}
- H04B 7/0684: {using different training sequences per antenna}
- H04B 7/06952: {Selecting one or more beams from a plurality of beams, e.g. beam training, management or sweeping}
- H04B 7/088: {using beam selection}
- H04B 17/328: {Reference signal received power [RSRP]; Reference signal received quality [RSRQ]}
- H04L 5/0051: {of dedicated pilots, i.e. pilots destined for a single user or terminal}
- H04B: TRANSMISSION
- G06N: COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- H04L: TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION (arrangements common to telegraphic and telephonic communication H04M)

## Abstract

A beam management method is provided. The beam management method is performed by a user equipment and includes receiving, by the at least one processor, reference signals corresponding to a transmitting beam by using a first receiving beam and a second receiving beam; identifying, by the at least one processor, a first reference signal receiving power (RSRP) of the first receiving beam and a second RSRP of the second receiving beam; analyzing, by the at least one processor, the first RSRP and the second RSRP respectively by a machine learning model; and identifying, by the at least one processor, a beam management operation corresponding to the transmitting beam according to a result of the analyzing.

## Claims

### Claim 1

A method performed by a user equipment (UE), the method comprising:
determining a receiving beam ratio of first receiving beams to a second receiving beam;
receiving reference signals according to the receiving beam ratio corresponding to a transmitting beam by using at least one of the first receiving beams and the second receiving beam;
identifying reference signal receiving power (RSRP) of the received reference signals;
analyzing each RSRP by a machine learning model; and
identifying a beam management operation corresponding to the transmitting beam according to a result of the analyzing,
wherein the first receiving beam covers a first range, and the second receiving beam covers a second range that is wider than, and comprises, the first range of the first receiving beam, and
wherein the receiving the reference signals comprises:
measuring the reference signal a preset number of times alternately using the first receiving beam and the second receiving beam according to a preset beam ratio while monitoring a rotation speed of the UE,
wherein the measuring of the reference signal changes the preset beam ratio and the preset number of times for measuring the reference signal based on the rotation speed of the UE.

### Claim 2 (depends on 1)

The method according to claim 1, wherein the receiving the reference signals comprises:
identifying the at least one first receiving beam and the second receiving beam based on a current beam pattern and a predefined selection rule;
receiving multiple reference signals through the at least one first receiving beam and the second receiving beam; and
identifying multiple RSRPs respectively corresponding to the multiple reference signals.

### Claim 3 (depends on 1)

The method according to claim 1, wherein a correspondence between the first receiving beam and the second receiving beam changes as an antenna array configuration of the UE changes.

### Claim 4 (depends on 1)

The method according to claim 1, wherein the first receiving beam comprises any one or any combination of an active beam and a predefined receiving beam of the UE.

### Claim 5 (depends on 4)

The method according to claim 4, wherein the receiving the reference signals further comprises:
identifying a receiving beam order of the first receiving beam and the second receiving beam;
receiving the corresponding reference signal according to the receiving beam ratio and the receiving beam order by using the first receiving beam and the second receiving beam sequentially; and
identifying the RSRP of the corresponding reference signal.

### Claim 6 (depends on 1)

The method according to claim 1, wherein the result of the analyzing indicates any one or any combination of:
a predicted value of a state of each receiving beam of the UE;
an evaluation value of each receiving beam of the UE;
a beam ranking corresponding to each receiving beam of the UE;
a candidate set of optimal receiving beams;
a long-term value when each receiving beam of the UE is an active beam;
the beam to be measured;
a measuring order; and
a beamforming codebook to be used.

### Claim 7 (depends on 1)

The method according to claim 1, wherein the method further comprises, before analyzing each RSRP by the machine learning model:
arranging multiple RSRPs according to a predefined format, the multiple RSRPs comprising the first RSRP and the second RSRP; and
performing data preprocessing on the multiple RSRPs, respectively.

### Claim 8 (depends on 7)

The method according to claim 7, wherein the performing data preprocessing comprises any one or any combination of the following actions:
increasing an offset; downscaling data of a first predefined ratio; upscaling data of a second predefined ratio; increasing an offset and downscaling data of the first predefined ratio; increasing an offset and upscaling data of the second predefined ratio.

### Claim 9 (depends on 1)

The method according to claim 1, wherein the machine learning model comprises a reinforcement-learning-based Q learning model, and
wherein a Q function of the reinforcement-learning-based Q learning model comprises an artificial neural network model and hidden layers of the reinforcement-learning-based Q learning model comprise a bidirectional long short term memory network model.

### Claim 10 (depends on 1)

The method according to claim 1, wherein a training method of the machine learning model comprises:
obtaining a predefined amount of original data;
processing the original data according to a predefined reinforcement learning method to obtain corresponding training data; and
training the machine learning model based on the training data to obtain a trained machine learning model.

### Claim 11 (depends on 1)

The method according to claim 1, further comprising:
performing corresponding beam management based on the beam management operation; and
refining corresponding parameters of the machine learning model based on a processing result of the beam management to further optimize the machine learning model.

### Claim 12 (depends on 1)

The method according to claim 1, wherein the identifying the beam management operation corresponding to the transmitting beam comprises any one or any combination of the following operations:
determining to switch the receiving beam;
switching a current receiving beam to an identified receiving beam, the identified receiving beam being identified based on a beam state and/or an evaluation value;
maintaining a current beam receiving configuration;
identifying whether to switch the current beam receiving configuration;
identifying a subsequent receiving beam for a subsequent measurement period based on the beam state and/or the evaluation value; and
identifying whether to switch a receiving beam and to change a beamforming codebook.

### Claim 13 (depends on 1)

The method according to claim 1, wherein the UE is operating in a carrier aggregation state where multiple carrier components are included, wherein the machine learning model comprises a corresponding machine learning model for each of the multiple carrier components,
wherein the receiving the reference signals comprises:
for each carrier component, receiving the reference signals corresponding to the transmitting beam by using the at least one first receiving beam and the second receiving beam corresponding to a corresponding carrier component, respectively, and identifying a corresponding RSRP of each received beam, and
wherein the analyzing each RSRP respectively by the machine learning model, comprises:
analyzing each RSRP respectively by the corresponding machine learning model.

### Claim 14 (depends on 13)

The method according to claim 13, further comprising:
identifying a beam pattern corresponding to each of the multiple carrier components based on a codebook of each of the multiple carrier components; and
identifying at least two receiving beams respectively corresponding to each of the multiple carrier components based on a corresponding beam pattern and a predefined beam selection rule.

### Claim 15 (depends on 1)

The method according to claim 1, wherein the UE measures multiple reference signals respectively transmitted on multiple transmitting beams,
wherein the receiving the reference signals comprises:
for any of the multiple transmitting beams, receiving the reference signals transmitted by the multiple transmitting beams by using at least two receiving beams, respectively, and identifying an RSRP of each receiving beam,
wherein the analyzing each RSRP respectively by the machine learning model, comprises:
for any one of the multiple transmitting beams, analyzing each RSRP by a corresponding machine learning model,
wherein the identifying the beam management operation corresponding to the multiple transmitting beams, comprises:
for any of the multiple transmitting beams, identifying the beam management operation corresponding to the multiple transmitting beams, and
wherein, the method further comprises:
performing a corresponding beam management according to the multiple transmitting beams based on transmission data being received from any of the multiple transmitting beams.

### Claim 16 (depends on 15)

The method according to claim 15, further comprising:
performing the corresponding beam management according to the beam management operation of the switched transmitting beam of the transmission data based on the transmission data being switched.

### Claim 17

A user equipment (UE), the UE comprising:
a transceiver; and
at least one processor configured to:
determine a receiving beam ratio of a first receiving beams to a second receiving beam;
receive reference signals according to the receiving beam ratio corresponding to a transmitting beam by using at least one of the first receiving beams and the determined second receiving beam,
determine reference signal receiving power (RSRP) of each receiving beam,
analyze each RSRP by a machine learning model, and
identify a beam management operation corresponding to the transmitting beam based on an output of the analysis result,
wherein the first receiving beam covers a first range, and the second receiving beam covers a second range that is wider than, and comprises, the first range of the first receiving beam, and
wherein the at least one processor:
receives the reference signals to measure the reference signal a predetermined number of times alternately using the first receiving beam and the second receiving beam according to a predetermined beam ratio while monitoring a rotation speed of the UE,
wherein the measuring of the reference signal changes the predetermined beam ratio and the predetermined number of times for measuring the reference signal based on the rotation speed of the UE.

### Claim 18 (depends on 1)

The method according to claim 1, further comprising identifying a receiving beam order based on whether the rotation speed exceeds the threshold,
wherein the receiving the reference signals is performed according to the receiving beam order.

### Claim 19 (depends on 1)

The method according to claim 1, wherein the at least one first receiving beam and the second receiving beam further comprises a third receiving beam covering a third range, and
wherein the second range of the second receiving beam comprises the first range of the first receiving beam and the third range of the third receiving beam.

### Claim 20

A method performed by a user equipment (UE), the method comprising:
receiving reference signals corresponding to a transmitting beam by using at least two receiving beams;
identifying reference signal receiving power (RSRP) of each receiving beam;
analyzing each RSRP by a machine learning model; and
identifying a beam management operation corresponding to the transmitting beam according to a result of the analyzing,
wherein the at least two receiving beams comprise a first receiving beam covering a first range, and a second receiving beam covering a second range comprising the first range the first receiving beam,
wherein the first receiving beam comprises any one or any combination of an active beam and a predefined receiving beam of the UE,
wherein the receiving the reference signals comprises:
measuring the reference signal N1 times alternately using the first receiving beam and the second receiving beam, wherein the first receiving beam is the active beam;
monitoring a rotation speed of the UE;
based on the UE detecting that the rotation speed exceeds a threshold, measuring the reference signal N2 times alternately using the first receiving beam and the second receiving beam, wherein the first receiving beam comprises multiple predefined receiving beams;
identifying whether a signal of the UE is occluded; and
based on the UE detecting that the signal is occluded, measuring the reference signal N3 times alternately using the first receiving beam and a corresponding second receiving beam, wherein a measuring ratio of the first receiving beam and the second receiving beam is a predefined ratio, and the first receiving beam comprises the multiple predefined receiving beams and active beams, and
wherein N1, N2, and N3 are all positive integers.