Method and device for selecting reception beam on basis of artificial neural network

Patent metadata

Field	Value
Publication number	US-11888236-B2
Application number	US-201917295623-A
Country	US
Kind code	B2
Filing date	Sep 26, 2019
Priority date	Nov 26, 2018
Publication date	Jan 30, 2024
Grant date	Jan 30, 2024

Abstract

Disclosed in various embodiments of the present invention are a method and a device, the device comprising: an antenna module configured to form a plurality of beams having different directions; and a processor operatively connected to the antenna module, wherein the processor is configured to select a partial reception beam from among a plurality of reception beams, measure the reception power of the selected reception beam, determine a transmission condition through an artificial neural network on the basis of the measured reception power, and determine a reception beam for a communication connection by using the artificial neural network corresponding to the transmission condition. Various embodiments are possible.

First claim

The invention claimed is:

1. An electronic device comprising:
an antenna module configured to form a plurality of reception beams in different directions; and
a processor operatively connected to the antenna module,
wherein the processor is configured to
select some reception beams from among the plurality of reception beams,
measure reception power of the selected reception beams,
determine a transmission condition through an artificial neural network, based on the measured reception power, and
determine a reception beam for a communication connection using an artificial neural network corresponding to the transmission condition.

2. The electronic device of claim 1 , wherein the processor is configured to select at least two reception beams, which are not adjacent to each other, from among the plurality of reception beams.

3. The electronic device of claim 1 , wherein the processor is configured to determine the transmission condition by inputting the measured reception power to the artificial neural network for determining the transmission condition.

4. The electronic device of claim 1 , wherein the transmission condition comprises at least one of no transmission beam, NLOS (non-line of sight), or LOS (line of sight), and
wherein the processor is configured to predict a reception beam using an artificial neural network corresponding to the transmission condition.

5. The electronic device of claim 4 , wherein the processor is configured
if the transmission condition is “NLOS”, to predict a reception beam using an artificial neural network corresponding to “NLOS”, and
if the transmission condition is “LOS”, to predict a reception beam using an artificial neural network corresponding to “LOS”.

6. The electronic device of claim 4 , wherein the processor is configured, if the transmission condition is “no transmission beam”, to perform a reception beam prediction process for a reception beam that is not selected from among the plurality of reception beams.

7. The method of claim 1 , wherein the processor is configured
if reception power of the selected reception beam is less than or equal to a minimum power, to determine the transmission condition to be “no transmission beam”, and,
if the reception power of the selected reception beam exceeds the minimum power, to determine the transmission condition to be “NLOS” or “LOS”.

8. The electronic device of claim 7 , wherein the processor is configured
if a difference in reception power between the selected reception beams is less than or equal to a reference value, to determine the transmission condition to be “NLOS”, and
if a difference in reception power between the selected reception beams exceeds the reference value, to determine the transmission condition to be “LOS”.

9. The electronic device of claim 1 , wherein the processor is configured to
predict a reception beam using the artificial neural network corresponding to the transmission condition,
determine success or failure of communication through the predicted reception beam, and
determine a reception beam for a communication connection, based on the determination result.

10. The electronic device of claim 9 , wherein the processor is configured
to predict an error rate of the predicted reception beam, and
if the predicted error rate is less than a threshold, to determine the predicted reception beam to be a reception beam for a communication connection.

11. The electronic device of claim 9 , wherein the processor is configured
to predict an error rate of the predicted reception beam,
if the predicted error rate is greater than or equal to a threshold, to determine that communication fails, and
to perform a reception beam prediction process for a reception beam that is not selected from among the plurality of reception beams.

12. The electronic device of claim 1 , wherein the processor is configured to
predict a reception beam using an artificial neural network corresponding to the transmission condition,
measure reception power of the predicted reception beam, and
determine a reception beam for a communication connection, based on the measured reception power.

13. The electronic device of claim 12 , wherein the processor is configured
if the reception power of the predicted reception beam is greater than or equal to a reference value, to predict an error rate of the predicted reception beam, and
if the predicted error rate is less than a threshold, to determine the predicted reception beam to be a reception beam for a communication connection.

14. The electronic device of claim 12 , wherein the processor is configured
if the reception power of the predicted reception beam is less than a reference value, to predict an error rate of a reception beam having the highest reception power among the reception beams whose reception power was measured, and
if the predicted error rate is less than a threshold, to determine the reception beam having the highest reception power to be a reception beam for a communication connection.

15. An operation method of an electronic device, the method comprising:
selecting some reception beams from among a plurality of reception beams in different directions from each other;
measuring reception power of the selected reception beams;
determining a transmission condition through an artificial neural network, based on the measured reception power; and
determining a reception beam for a communication connection using an artificial neural network corresponding to the transmission condition.

Assignees

• Samsung Electronics Co Ltd

Inventors

• Cho Yongsang
• Kim Taeyoon
• Lim Chaeman
• Lee Hyoungjoo

Classifications

• G06N3/0499

  Feedforward networks · CPC title

• G06N3/09

  Supervised learning · CPC title

• H01Q3/2605 Primary

  Array of radiating elements provided with a feedback control over the element weights, e.g. adaptive arrays · CPC title

• G06N3/02

  Neural networks · CPC title

• H01Q3/24

  varying the orientation by switching energy from one active radiating element to another, e.g. for beam switching · CPC title