Granted Patent B2

US 12,446,035 · App. 17/896,659 · Granted Oct 14, 2025

Inventors: Satya Kumar Vankayala (Andhra Pradesh, IN); Venkateswarlu Yarramala (Andhra Pradesh, IN); Seungil Yoon (Suwon-si, KR)

Assignee: SAMSUNG ELECTRONICS CO., LTD.

H04W72/1273H04L1/0061H04L1/1812H04W72/23

Abstract

Claims (54)

1. A method of operating a user equipment (UE) in a wireless network for managing physical downlink control channel (PDCCH) data, the method comprising:

obtaining a plurality of network parameters;

before searching for a PDCCH candidate, predicting at least one aggregation level (AL) used by a base station (BS) associated with the UE to transmit the PDCCH data in the wireless network based on the plurality of obtained network parameters; and

decoding the PDCCH data based on the at least one predicted AL.

2. The method as claimed in claim 1 , wherein the predicting of the at least one AL based on the plurality of network parameters comprises:

providing the plurality of network parameters as an input to a trained machine learning (ML) model, and

obtaining the at least one predicted AL as output from the ML model.

3. The method as claimed in claim 2 , wherein the ML model includes a neural network (NN) associated with the BS trained using a dataset including a plurality of network parameters and a corresponding AL used by the BS to transmit PDCCH data.

4. The method as claimed in claim 2 , wherein the method comprises:

receiving a decoding feedback using the at least one predicted AL; and

updating the ML model based on the received decoding feedback.

5. The method as claimed in claim 1 , wherein the decoding of the PDCCH data based on the at least one predicted AL comprises:

searching a PDCCH candidate for the at least one predicted AL; and

detecting the PDCCH data from the searched PDCCH candidate based on the searched PDCCH candidate passing a cyclic redundancy check (CRC).

6. The method as claimed in claim 1 , wherein the decoding of the PDCCH data based on the at least one predicted AL comprises:

selecting an AL from the at least one predicted AL;

searching a PDCCH candidate for the selected AL; and

detecting the PDCCH data from the searched PDCCH candidate based on the searched PDCCH candidate passing a cyclic redundancy check (CRC).

7. The method as claimed in claim 1 , wherein the plurality of network parameters comprises at least one of current channel conditions, a hybrid automatic repeat request (H-ARQ) feedback, and a downlink control information (DCI) grant size.

8. The method as claimed in claim 1 , wherein the PDCCH data is decoded using a greedy mechanism, wherein the greedy mechanism is configured to:

start searching PDCCH candidates in the at least one predicted AL, and

upon finding a PDCCH payload, stop the searching and not proceed with searching remaining PDCCH candidates in the at least one predicted AL.

9. The method as claimed in claim 1 , wherein the method further comprises:

predicting at least one search space set to be monitored for the PDCCH data based on the plurality of obtained network parameters; and

decoding the PDCCH data based on the at least one predicted search space set.

10. A method of operating a base station (BS) in a wireless communication network for managing physical downlink control channel (PDCCH) data, the method comprising:

obtaining a plurality of network parameters;

predicting an aggregation level (AL) to be used for transmission of the PDCCH data to a user equipment (UE) in the wireless network based on the plurality of network parameters by at least:

providing the plurality of network parameters as an input to a trained machine learning (ML) model, and

obtaining the at least one predicted AL as output from the ML model;

encoding the PDCCH data based on the at least one predicted AL; and

transmitting the encoded PDCCH data to the UE.

11. The method as claimed in claim 10 , wherein the plurality of network parameters comprises at least one of current channel conditions, a hybrid automatic repeat request (H-ARQ) feedback, or a downlink control information (DCI) grant size.

12. The method as claimed in claim 10 , wherein the ML model includes a neural network (NN) associated with the BS trained using a dataset including a plurality of network parameters and a corresponding AL used by the BS to transmit PDCCH data.

13. A user equipment (UE) configured to manage physical downlink control channel (PDCCH) data in a wireless network, the UE comprising:

a memory;

a processor comprising processor circuitry; and

a PDCCH decoding controller comprising processing circuitry, communicatively connected to the memory and the processor, configured to:

obtain a plurality of network parameters,

before searching for a PDCCH candidate, predict an aggregation level (AL) used by a base station (BS) associated with the UE to transmit the PDCCH data in the wireless network based on the plurality of obtained network parameters, and

decode the PDCCH data based on the at least one predicted AL.

14. A base station (BS) configured to manage physical downlink control channel (PDCCH) data in a wireless network, the BS comprising:

a memory;

a processor comprising processor circuitry; and

a PDCCH encoding controller comprising processing circuitry, communicatively connected to the memory and the processor, configured to:

obtain a plurality of network parameters,

predict an aggregation level (AL) to be used for transmission of the PDCCH data to a user equipment (UE) in the wireless network, based on the plurality of network parameters, by at least:

providing the plurality of network parameters as an input to a trained machine learning (ML) model, and

obtaining the at least one predicted AL as output from the ML model,

encode the PDDCH data based on the at least one predicted AL, and

transmit the encoded PDCCH data to the UE.

15. The user equipment (UE) as claimed in claim 13 , wherein the PDCCH encoding controller is further configured to predict the at least one AL based on the plurality of network parameters by:

providing the plurality of network parameters as an input to a trained machine learning (ML) model, and

obtaining the at least one predicted AL as output from the ML model.

Assignments (1)

Priority Claims (2)

Continuity (2)

References Cited (22)