US 11,855,657 B2
Method and apparatus for decoding data packets in communication network
Tirthankar Mittra, Karnataka (IN); Seungil Yoon, Suwon-si (KR); and Satya Kumar Vankayala, Karnataka (IN)
Assigned to Samsung Electronics Co., Ltd., Suwon-si (KR)
Filed by SAMSUNG ELECTRONICS CO., LTD., Suwon-si (KR)
Filed on Jul. 1, 2022, as Appl. No. 17/856,004.
Application 17/856,004 is a continuation of application No. PCT/KR2022/008528, filed on Jun. 16, 2022.
Claims priority of application No. 202241017089 (IN), filed on Mar. 25, 2022.
Prior Publication US 2023/0308116 A1, Sep. 28, 2023
Int. Cl. H03M 13/11 (2006.01); G06F 18/214 (2023.01)
CPC H03M 13/114 (2013.01) [G06F 18/214 (2023.01); H03M 13/1108 (2013.01); H03M 13/1122 (2013.01); H03M 13/1128 (2013.01); H03M 13/1177 (2013.01)] 17 Claims
OG exemplary drawing
 
1. A method for decoding data packets by an apparatus including a processor, in a communication network, the method comprising:
pre-training, by the processor, a reinforcement model;
receiving, by the processor, one or more data packets related to each of one or more data types; and
decoding, by the processor, the one or more data packets using a parity check matrix associated with a corresponding data type, wherein the parity check matrix comprises a plurality of layers, arranged according to a combination of layers which is determined using the pre-trained reinforcement model,
wherein pre-training the reinforcement model comprises:
receiving, by the processor, training data packets related to each of a plurality of training data types;
identifying, by the processor, a pre-set parity check matrix for the training data packets related to each of the plurality of training data types;
decoding, by the processor, the training data packets related to each of the plurality of training data types using each combination of layers of a plurality of combinations of layers of the pre-set parity check matrix;
measuring, by the processor, a decoding time associated with decoding of the training data packets related to each of the plurality of training data types for each combination of layers; and
identifying, by the processor, one or more combination of layers having a minimum decoding time for each training data type based on the measured decoding time.